Biology-based Philosophy and Science Practice

2 downloads 0 Views 485KB Size Report
Science or science fiction: beyond practice frontiers ..... influence the organisation of permanently moving solar systems or galaxies. ...... Visible stars and their.
1

Biology-based Philosophy and Science Practice Four Decades of Observation and Interpretation Marcel (M.) Lambrechts Centre d’Ecologie Fonctionnelle et Evolutive, CEFE UMR 5175, Campus CNRS, 1919 route de Mende, Montpellier Cedex 5 F-34293, France Tel.: 33 (0) 4 67 61 33 10, FAX: 33 (0) 4 67 41 21 38 E-mail: [email protected] DOI: 10.13140/2.1.3456.7048 ‘Humans worldwide produced billions of written pages, oral sentences or thoughts in the past. There is therefore no doubt that remarks or questions listed below have been expressed by others one way or another. Citing published sources dealing with the same or similar topics would create additional bias and useful or useless discussion. Sources at the origin of ideas are difficult to identify given the complexity of visible or invisible information exchange.’

22 August 2014 The perceived organisation of the world differs amongst living beings, being ants, birds, humans and, why not, extra-terrestrial organisms. Citizens definitely agree that non-human vertebrates with biased biology will not be able to communicate elaborated human philosophy products. Physiologists or neurobiologists will probably claim there will be no human philosophy in the absence of biological structures that store information or create ideas. Evolutionary biologists might claim that human brains have been selected by natural or sexual selection to favour survival or reproduction not requiring the perception of all the details in nature that are studied in an empirical research framework. A biology-based shift in perception within or across individual living beings might transform perceived ‘hazard’ into perceived ‘determinism’ or translate a perceived ‘random’ process into a perceived ‘deterministic’ process. Human philosophy and science practice, also guided by daily life experiences and thus empirical observation, could be defined as examples of expressed behaviour influenced by a cocktail of underlying biology-based mechanisms. The empirical approach in science is principally based on analysis of perceived patterns (e.g. visual, sound) reflected from or produced by ‘matter’, like material, soil, rock, crystal, liquids or any living being. Sophisticated science tools, such as electron microscopes, receptors, spectrometers, telescopes, cameras, recorders, binoculars or data loggers will always observe ‘matter’ from a distance. Because physics of reflectance patterns of light or any transmitted form of energy from ‘matter’ apparently differs from physics of ‘matter’ per se, empirical observation from any distance will not give full access to matter physics per se. Thus empirical science observation would suffer from bias due to methodology based on perception not giving full access to fundamental physical matter structures in nature. Science is also principally based on believing what other people tell or write. There exists no fundamental difference in the practice of science and religion when both accept existence of empirically unverifiable theory or both believe in existence of phenomena claimed to be perceived by a minority of the human community. Scientists are apparently more convinced when phenomena are empirically accessible to a community of observers. But how often does this happen from an empirical point of view that efforts to replicate research activity are not supported by science commissions permanently searching for novelty? People or research

2

teams often claim they observe unique phenomena, appearances or sensations that cannot (always) be verified or replicated. For instance, science uses simple measures of brain activity to study sleep patterns reflected in encephalograms or receptor responses or brain scans. However, these measurements are insufficient to reveal the complex visual images, stories or senses perceived during sleep. Most, if not all, citizens claim dreams exist, but scientists can currently not retrieve or replicate complex dynamic dream images or movies to demonstrate their existence other than quantifying what (lucid) dreamers tell. Individuals practicing philosophy or science are thus confronted with bias in individual perception and reasoning related to biology-based constraints and the rising tide of the information (data, methods, literature) deluge in a continuously changing world. Individuals might therefore be perceived by others as ‘expert’, ‘generalist’, ‘competent’ or ‘incompetent’ for at least one human-defined scale of analysis. Fidelity to a given research field or model species or system might increase scientific visibility and credibility from a practical point of view when it ultimately results in democratic science consensus (‘DSC’) amongst people willing to tell the same or similar stories. Without ‘DSC’ supported by a critical mass of people there will never be scientific or social credibility towards potential believers, like citizens, politicians or decision-makers that influence socioeconomic or cultural activity. Many resources are invested in research aimed to convince ‘non-believers’. Perhaps intelligent non-believers will always be able to find arguments not accepting proposed interpretations of research findings from others. They might philosophically argue that scientific hypotheses can never be rejected or accepted. Perhaps individual prestige, personality, mental state, individual perception constraints, social integration or professional external pressures are at the origin of educated people rejecting reasonable constructive propositions or well-argued critical remarks. The appropriate scale of analysis of scientific problems is often not fully understood. Experts might have one scale of analysis or perception in mind whereas citizens might think about other scales. Thus, what philosophers or scientists observe and learn will depend on who they are and where they are causing different individuals to communicate or perceive similar or different messages or stories. The challenge will then be to define scales of analysis and perception potentially accessible to different science practisers. This also includes accepting existence of scales currently not accessible from an empirical or practical point of view. This will be exemplified in more detail below using both hypothetical case studies and empirical observation or interpretation in scientific decision-making. The unanswered questions and remarks listed below are aimed to stimulate discussion and thinking. Science or science fiction: beyond practice frontiers Philosophers or scientists ultimately aim to define unifying theories to explain as much as possible numerous phenomena based on a few principles. Despite the limits imposed by biology, philosophers think they can explore topics inaccessible to current science practice. For instance, philosophers following Russell might postulate that two phenomena separated in time or space always differ in at least one scale of analysis or perception. A scientifically unverified, but philosophical acceptable, statement is that ‘each studied phenomenon is unique in (human-) perceived physical expression’. Combined effects of Rutherford and Heisenberg principles might cause each particle or energy wave to be unique in measurable or immeasurable physical expression. Physical bodies that combine particles or energy waves at higher levels of organisation should therefore also be unique in expression. Thus, two perceived grains of sand on a beach or two individual oxygen atoms would never be structurally exactly the same at all scales of analysis. Moreover, because phenomena will

3

probably be perceived differently by organisms with different biology, philosophers might also state that for at least one scale of analysis two phenomena will never be perceived exactly the same by different observers. This would imply that physical, chemical or biological structures are physically always 'un-replicable' or perceived as ‘un-replicable’. Scientifically 'replicable' phenomena might therefore be not more than human mental products disconnected from the true nature of nature. Science practisers would thus be forced to accept imprecisions when they define classes or groups of phenomena with common characteristics. They might invent human-created rules to make empirical science practice feasible for at least one humandefined scale of analysis or perception. The greatest scientific challenge might therefore be to demonstrate that two phenomena separated in time and space are physically exactly the same at all scales of analysis and perception, also to justify the usefulness of what scientists name ‘true replication’ in empirical science practice. Daily experience in citizens: dreaming as model system Daily experience and its memorisation can construct huge sets of personal data stored in the individual brain. These personal data are often not accessible to other individuals, like scientists. It represents a biology-based individual citizen’s data set available for individual interpretation that influences individual decision-making. Can daily life provide sufficient data and inspiration to propose potentially interesting topics currently unfamiliar in science practise? As an example, both citizens and scientists might link dream expression to experienced internal states or external factors. Citizens might compare personal dream experiences with those of other people telling the same or different stories. Both citizens and scientists might come to the conclusion that dream contents apparently differ between professions or cultures. Dream scientists claim that dream expression would only be based on recombination of stimuli perceived and stored when people are awake. Dream scientists also claim that they cannot predict which dream images will be expressed in experimental conditions or that dream stories cannot be fully controlled by dreamers. However, how to explain why citizens occasionally perceive exactly the same series of dream images with intervals of several weeks. The probability that dream copies are expressed in controlled laboratory conditions is probably quite low. Some citizens also claim they dream about sharp colourful images of human faces never met before and perceived again during a meeting in ‘real’ life a couple of days after dream images were expressed. Can this phenomenon be scientifically demonstrated in laboratory settings? Given there are billions of dream expressions per night in the world, only one convincing case study might be sufficient to show that dream expression can also be based on external environmental stimuli received via the sensory system during sleep or non-sleep states without biologically active eyes. Perhaps some aspects of dream expression and perception follow similar physical laws as those from television or radio signals transmitted from distant sources and ultimately expressed in receivers. Can it happen at least once in a lifetime that dream images are simultaneously shared amongst people in experimental laboratory settings? Can scientists fully exclude that energy waves from external sources are unconsciously received by the human brain and subsequently expressed in dream images? Mental or physiology states of individual living beings are definitely influenced by energy transmitted from distant material or objects, such as devices producing electromagnetic fields. Specialists in hypnosis apparently control actions and mental states in those that are hypnotised, even at a distance in a theatre. It implies that external factors can rapidly change individual mental perception states potentially creating rapid shifts in how the external environment is perceived by individual living beings. Citizens talking about personal

4

experiences might inspire scientists to explore unfamiliar fields, but when do scientists start to believe unfamiliar stories from citizens. New knowledge might be created through the recombination of existing knowledge. If dreams result from recombination of perceived information to create novel visual images, dreams might be drivers of cultural novelty and evolution. Dream expression might also be considered as a form of social communication or might result from visible or invisible communication networks when dreams are shared amongst dreamers. People might be more attracted to persons that provoke pleasant dreams. Thus dreams expressed at night might influence how people behave or communicate during the day. Perhaps dreams are amongst the cheapest forms of social travelling and communication. Do individual dream scientists use personal experience to interpret dream stories provided by others in dream experiments? Would scientists have accepted the incredible dream stories told by citizens if they would not have had the same experiences themselves? The scientifically required details of individually experienced dreams can currently not be recorded and visualised on a screen or printed on paper, which is evidently also the case for feelings and motivations of which a biased part is reflected in oral expression, gestures or art. There is evidently a substantial difference between a film and the description of a film perceived.

Differences between theory and practice Philosophers might claim that hypothesis-testing is limited to approaches principally aimed to reject hypotheses. Searching for what scientists name ‘evidence’ would therefore not be an adequate approach. Perhaps many discoveries that significantly contributed to the functioning of the human world did not result from philosophically recommended practice. Philosophers might also argue that hypotheses can never be fully rejected because of biology-based constraints causing observation or perception bias, whatever the complexity of techniques applied. Perhaps true scientific challenge in an empirically defined world is to identify scales that are accessible to more than one human observer. Many research disciplines like Ecology, Ethology or Psychology are considered ‘soft’ sciences scientifically not truly ‘exact’ or ‘precise’ because of perceived complexity of phenomena investigated. But how exact or precise is science in general from an empirical or scientific or philosophical point of view? That empirical observation always differs from theoretical prediction can be illustrated with simple examples. How many figures after the comma are required to match theoretical prediction concerning dimensions of simple geometric figures? Who decides on the precision of measurements required? In other words, how exact is ‘exact’ or how ‘precise’ is ‘precise’? Moreover, simple geometric equations describe triangles but how well do the theoretical predictions provided by the ancient Greeks match practice? If children or adults are asked to measure the same triangle drawn on paper, and they apply mathematical equations describing surfaces or perimeters of that triangle, there is definitely an observer effect. Deviations from human-created theory may be caused by different factors. Measurement precision may change with the thickness of the lines constituting the triangle. Thicker lines may increase imprecision in measurement. The environment at the time of measurement influences perception therefore also determining how triangles are measured. More ambient noise may lower individual mental focus perhaps having impact on how triangles are perceived and measured in a class room. Evidently, precision of measurement will depend on material used. Thus, empirical measurement of the same triangle provides different results amongst observers even after controlling for

5

support and environment. If the true nature of nature is indeed variation, human-invented ‘perfect’ triangles as defined in Mathematics can never be identified with empirical science practice. Why should human-invented theory be right and human practice wrong? Who decides that human-invented theory not taking natural principles of variation into account is right? If one takes the reality of natural diversity into account measurements, not theory, are true. Alternatively, both human practice and human-invented theory might be considered true accepting all human products and activities are products from nature. If some people claim that the true scientific challenge is to reject existing hypotheses, how can this approach be compatible with motivations to discover to explain why phenomena exist? To demonstrate whether a theory, concept or hypothesis is true requires one convincing example supporting it, accepting principles of human-defined ‘DSC’. The scientific community may think that exact biological replicates (e.g. clones) exist for at least some empirical scales of analysis. However, as explained above, philosophers may argue that each individual living being is unique in physical structure and replicates that share exactly the same physical expression in every detail therefore would not exist. However, scientists frequently observe and verify to demonstrate or track dynamic phenomena, not to reject their existence. This is permanently done in medicine. Living beings could be described as automatized pilots of vehicles, named bodies. Living beings function without been aware of technical details of body processes involved. This allows the ‘conscious’ brain scanning the environment with senses to focus on and rapidly adjust decisions to changes in a small subset of dynamic external factors that influence individual living conditions. The conscious brain been aware of aspects in the external environment through senses may also feel and thus perceive changes in functioning of automatized body processes. Medical doctors discover or identify changes in automatized body processes of patients, eventually to repair them. Psychologists or behavioural specialists have been formed to discover and identify unusual changes in non-automatized (conscious) processes. Automatized body processes may evidently influence non-automatized processes, and vice versa. Thus medical doctors continuously verify and compare functioning of individual human bodies. They occasionally use advanced medical techniques to make a diagnosis. Exchange between patients and medical doctors will obviously contribute to the identification of underlying causes of patients’ feelings. Philosophers may argue that science practice is most often not truly scientific, but science can work from a practical point of view not taking philosophical arguments into account. Scientists invent techniques that improve the quality of life. Medical doctors provide concrete solutions for medical problems. Thus medical doctors and patients do not have access to theoretical or philosophically defined details of the functioning of the human body, but they are able to live, survive and reproduce in a world they perceive from a practical or empirical point of view. In general, mankind does not have to understand or perceive every detail in nature to live and survive. The whole world could be defined as a complex automatized system of which only a tiny fraction will ultimately be perceived an exploited by each individual living being.

Science terminology and philosophy Scientists or philosophers may propose that hypotheses that cannot be empirically tested or verified should not be exposed or published. Philosophers or scientists may also argue that science should only propose terminology that can be empirically tested or verified with ‘precision’. ‘Mathematical Symmetry’ or rotating and discrete symmetry transformations as defined in pure Mathematics cannot be identified with an empirical approach when 'perfect

6

circles’ or ‘perfect squares’ cannot be drawn and measured because of biology-based constraints, such as constraints preventing human observers to perceiving every detail in nature. If ‘Mathematical Symmetry’ cannot be observed or identified in practice should 'Symmetry' simply be replaced by empirically measurable terminology, like relative 'Deviation', 'Variation', 'Difference', ‘Asymmetry’? And what about other terminology like ‘Fixed’, ‘Static’, or ‘Invariable’ that can only be defined accepting imprecisions in at least some empirical scales of analysis? Evidently, ‘absolute’ values philosophically require an infinite amount of numbers behind the comma (A = 1, etc. or B = 2, or A=B), and is therefore empirically less or not accessible compared to ‘relative’ differences (AB). If science terminology is not precise or detailed enough, how far can philosophers go with the proposition of examples that help to define terminology? This can be illustrated with a definition of what scientists name ‘phenotypes’ and ‘extended phenotypes’. Dawkins' published definition of phenotypes is ‘the manifested attributes of an organism, the joint product of its genes and their environment during ontogeny. A gene may be said to have phenotypic expression in, say, eye colour. …the concept of phenotype is extended to include functionally important consequences of gene differences, outside the bodies in which the genes sit.’ Dawkins’ published definition of extended phenotypes is ‘All effects of a gene upon the world. As always, 'effect' of a gene is understood as meaning in comparison with its alleles. The conventional phenotype is the special case in which the effects are regarded as being confined to the individual body in which the gene sits. In practice it is convenient to limit 'extended phenotype' to cases where the effects influence the survival chances of the gene, positively or negatively.’ Thus phenotypes can proximately be defined as components that physically belong to the body or so-called gene vehicle (e.g. legs, physical aspects related to body colours, organs or body size). Often cited examples of extended phenotypes are animal artefacts physically expressed outside the body in the presence or absence of the body, such as nests or houses. Animal constructions have a genetic or environmental basis possibly also influencing processes in natural or sexual selection. However, some biological traits described as bodily phenotypes in scientific publications could perhaps also be defined as extended phenotypes. Physical expression of extended phenotypes differs from physical expression of phenotypes producing physical expression of extended phenotypes. For instance, the physical phenotype of a sender, such as a body of a given physical size, is translated into an ‘extended phenotype’ when a light reflectance pattern from a body of a given physical size is transported in different kinds of medium (e.g. air, water) to be perceived as ‘large’ or ‘small’ depending on distance between senders and receivers. Bodily phenotypes are therefore not ‘fixed’ phenotypes from a perception point of view. Moreover, extended phenotypes can be expressed in the absence of phenotypes that produced extended phenotypes, like nests or castles physically present in the absence of builders. Thus, sound received from space today and produced thousands of years ago by currently extinct bodies are extended phenotypes. Bird songs or other vocalisations produced by the ‘invisible’ bodily phenotype hidden behind a bush are also extended phenotypes. Physics of the vocal organ (size) impose constraints on how energy perceived as song is produced and transported in a medium. The perceived sound pitch produced by a large vocal organ differs from that of a small one. There is a correlation between the physical structure of the phenotype and extended phenotype in at least one perceived scale of analysis. Do perception mechanisms of receivers translate perceived energy produced by the vocal organ of a sender and transmitted through air or water into sound? Some people claim they hear music or speech in the absence of a physical body producing sound. Does music per se exist or is it not more than physical energy expressed in

7

how moving air is perceived by the ear and translated into ‘music’ by the perception mechanism? Evidently, the same extended sound phenotype will be perceived differently by different living beings differing in spatial position or physics of perception mechanisms. Thus the perceived genetic or environmental basis of reflectance or sound patterns results from physical actions and interactions between senders and receivers. Moreover, transported odour or expiration molecules evidently are physically disconnected from the physical structure of organs or cells producing molecules (e.g. gland, lung, or air tract) and they can all defined as extended chemical phenotypes. Eye colour defined as a bodily phenotype could also be defined as an extended phenotype when they represent perceived reflectance patterns of which reflectance physics differs from that of physics of the biological eye. Physical structure of reflectance patterns might have a genetic or environmental basis, such as the quantitative genetics of reflectance patterns in avian plumages. Reflectance patterns could result from ecological or evolutionary action expressed in natural or sexual selection. This is the case when plumage reflectance patterns have an impact on mate choice, social interactions or organism-environment interactions that influence contributions to following generations. However, the same light reflectance patterns transmitted by a distant body will be perceived differently by different living beings when physics of brain and neuron systems express variation across receivers. Thus, extended phenotypes from bodies could be light waves, heat or other types of energy reflected or produced by physical action and transported through different types of medium (air, water). Perception mechanisms of living beings (brain, neurons) will transform received energy of distant ‘matter’ into exploitable information to be used in social communication or ‘organism – environment’ interactions. A perception mechanism is here considered as a filter of energy of which some energy is translated into colour perception, sound perception, etc.... and other energy not ‘consciously’ perceived. Empirical science practice based on observation and analysis of reflectance patterns or energy waves evidently study phenomena from the past because there is always a difference between the onset of emission (e.g. the start of light reflectance) and the onset of ‘conscious’ perception, whatever the distance between sender and receiver and the observation techniques used. Any scientific empirical observation from any distance (> 0.000000000...1 mm) of any ‘matter’ would therefore only give access to the study of extended phenotypes energetically disconnected from bodily phenotypes. Philosophy and terminology in ecology or biology Is terminology in science truly 'precise' to capture the true nature of nature? This can be illustrated with following examples in Biology or Ecology. What is the exact definition of 'reproduction'? To remake/copy/replicate what already exists? Philosophers might claim that each newly produced individual is unique in physical expression and is therefore not reproducible. This would imply that ('exact') reproduction does not exist given the physical dynamics of nature or that the details of physical structure cannot be used to define what reproduction is. . What is the exact definition of 'survival'? To maintain what already exists? Philosophers might claim that each newly produced biological unit is unique in physical expression at each spatiotemporal moment. This would imply that (exact) survival does not exist or is closely related to human-invented time given the continuous physical dynamics of nature.

8

What is the exact definition of 'restoration'? To remake/copy/replicate what existed in the past? Philosophers might claim that each biological unit, including an ecosystem or a component of an ecosystem, is unique in physical expression at each spatiotemporal moment. This would imply that ('exact') restoration does not exist given the physical dynamics of nature. Do imprecisions in science terminology result from the fact that scientists are not language specialists? Should science terminology result from interactions between scientists and language specialists? How detailed should science language be? Science practice accepts imprecisions in how phenomena are described. The imprecisions are reflected in words that classify phenomena together of which we assume they have characteristics in common (interpretation). The problem is that each phenomenon (object, event, living organisms) is unique in physical expression. For instance, different phenomena that we call 'nest' do not share exactly the same physical structure. Each structure what we name ‘nest’ is unique in physical expression. Each nest therefore could potentially be given a different (scientific) name (e.g. nest A, nest B, nest C, etc...) substantially increasing the complexity of the language used to describe phenomena in nature. Is there coherence in shared rules concerning details in language expression across scales of analysis? Providing individual-specific names to individual phenomena observed in nature will significantly increase language complexity. We give individual names to different planets or different suns (of the same class) or different galaxies, thus natural phenomena that can be observed for (thousands of) years/centuries. We don't give individual names for different individual bird nests, thus natural phenomena observed during brief periods of time (e.g. a couple of weeks). Has there been scientific consensus about how long individual phenomena must persist in time before they are accepted to be named individually? And what is physical persistence in time given that each (dynamic) phenomenon is unique in physical expression at any spatiotemporal moment?

Practical versus theoretical frontiers defining science disciplines Theoretical and practical frontiers of science disciplines might change when specialised research is replaced by multi-disciplinary, inter-disciplinary, pluri-disciplinary or transdisciplinary research that introduces novel ideas or new techniques in a more or less coordinated manner. How a research domain is theoretically defined or framed may influence how it is practically perceived by citizens or experts from other research domains. This can be illustrated with following hypothetical examples. Scales of analysis in studies of physical forces Specialists studying Quantum Physics claim they study Fundamental Laws of Nature when they study ‘strong’ nuclear forces or ‘weaker’ electromagnetic forces that involve basic building stones in nature. Specialists studying Chemistry claim they study Fundamental Laws of Nature when they examine forces that combine elements (atoms) that construct molecules. Astrophysicists claim they study Basic Laws of Nature when they focus on gravity forces that influence the organisation of permanently moving solar systems or galaxies. Some experts in Physics or Chemistry apparently are desperately searching for unifying theories that take attractive and repulsive forces at very small (Quantum Physics) and very large spatiotemporal

9

scales (Astrophysics) into account. They occasionally construct expensive ‘science cathedrals’ to study ‘invisible’ phenomena. But why do specialists in Quantum Physics or Astrophysics ignore forces at intermediate spatiotemporal scales of analysis in a discipline perhaps to be called Biophysics? Numerous scientists support the idea that living beings are fundamentally built up from energy waves, energy particles, atoms and molecules studied in Physics and Chemistry. Biophysics could perhaps be defined as a discipline aimed to better understand attractive or repulsive forces amongst living beings that compose mates, groups, species or communities. Why two individuals exposed to the ‘same’ perceived earth gravity force decide to approach or avoid each other? How do physically ‘invisible’ fields influence individual decisions concerning initiation of attraction or avoidance? Are ‘biotic social’ forces that act at the scale of Biophysics disconnected from ‘abiotic physical’ forces that act at the scale of Astrophysics or Quantum Physics? Some philosophers assume that every aspect in nature could be defined as Physics. Phytotherapy ‘Phytotherapy’ is often defined as ‘treatment with plants or plant components’. Following recent estimates of ‘O.M.S.’ (270.000 plant species currently identified in taxonomic Botany. Pharmaceutical treatment involves digestion or physical penetration of molecules or plant components targeting physiological or neurophysiological processes. However, psychologists guiding depressed people might argue that flowers or any vegetation used to decorate gardens, houses or green space also involve ‘Phytotherapy’. Perception of green space, including park or forest landscape, can reduce mental stress therefore improving health status. Moreover, diet specialists argue that consumed fruits contain all ingredients essential for physiological and, why not, mental performance. Thus vegetable consumption could be defined as ‘Phytotherapy’ when vegetables form an essential part of the human diet that improves health status. The question then is why those that defined ‘Phytotherapy’ made a human-invented distinction between plants consumed or digested versus plants observed or perceived? Plants digested and plants perceived are just acting on different parts of the human body, either via physiology (energy, metabolism) or neuron system receptors. Any plant component that improves health status could thus by definition be placed under the heading ‘Phytotherapy’. Thus, phytotherapy would involve all visual vegetation stimuli or components that lower stress and consequences of stress (e.g. depression), all vegetation stimuli or components that act directly on physiology, neurology or immunology, or all plants or plant substances that change external appearance to improve social interaction and integration. It does not exclude that actions of vegetation improving health are individual-specific, whatever the plant components involved. Some people might consider wildlife herbs as waste to be removed from urban green space, whereas others might consider the same wildlife herbs as essential parts of the urban landscape that reduce mental stress. Some people might also be more or less allergic to one plant component, even those currently considered as treatment in pharmaceutical research. This would imply that the definition of ‘Phytotherapy’ is individual-specific. Nearly all plants may have phytotherapeutic action in at least one individual, just depending on scales of analysis and application defined. Animal science: does it exist? According to the published literature, animal art exists. What about animal science? If the principles of (empirical) science practice are universal, perhaps ants apply the principles of empirical science practice, like 'trial and error', eventually selecting the best solution to solve a problem based on past experiences. Primates use tools to find food and transmit the methods

10

of tool use to others. Many living beings apparently apply basic principles of science theory correctly without necessarily understanding the theory or underlying mechanisms? How often reflects 'human science' 'animal science'. For instance, how often do scientists truly understand the underlying mechanisms of the patterns they describe when they apply trial and error approaches in empirical research?

Scale-dependent causality Science practice is assumed to be based on causality because what is studied (e.g. phenomenon, object or organisms) will result in biology-based perception and mental effects that change behavioural or physiological expression in science practisers. When expression in a phenomenon ‘A’ causally determines expression in a phenomenon ‘B’ that is physically nearby, a change in ‘A’ should according to human logic temporally be followed by a change in ‘B’. The types of interactions between ‘A’ and ‘B’ will determine how a change in ‘A’ may contribute to a change in ‘B’. For instance, when a person touches a ball on a table the ball moves through physical action. However, the ball will also move when the person lifts the table without touching the ball. Is movement of the ball caused by the position of the table surface or the physical action of the person lifting the table? The person can lift the table differently causing the ball to move differently. The structure and position of the table surface being in contact with the ball and the physical structure of the ball will by human definition cause direction and speed in movement of the ball. Events or phenomena are apparently interconnected more or less through a chain reaction involving other events or phenomena. Phenomena or events apparently belong to an interconnected network of multiple events and phenomena. A change in one component might cause a chain reaction causing other components of the network to change more or less. According to human logic, strong interconnected causality connecting ‘C’ to ‘P’ implies that a change in ‘C’ will provoke a change in ‘P’ temporally following ‘C’, although the expression of ‘P’ might be determined by other causes that reinforce or reduce effects of ‘C’ on ‘P’. How and when phenomena or events are interconnected and perceived will depend on how the causal network is structured and analysed by the analysers involved. A single phenomenon may thus be influenced by a chain reaction that involves numerous factors that determine its final expression. Specialists from different research domains may identify different causes for the same phenomenon depending on what aspect of the chain reaction is studied. A conspecific song 'S' at spot 'A' may attract a bird 'B' to spot 'A'. Song 'S' causes bird movement towards spot 'A'. A specialist in Neurobiology may think that it is the perception of the song that caused the bird to move. A specialist in Anatomy could propose that the bird moved because it has functional wings or legs. A specialist in Psychology might propose that it is the feeling caused by the perception of the song that caused the bird to move towards spot ‘A’. A specialist in Geography might think that it is the position of the song at spot ‘A’ that caused the direction of bird movement, etc. Thus, a phenomenon might result from a cocktail of causal factors or processes that evidently occurred at different moments and places in the past, accessible or not accessible to the human observer. All these aspects can be studied from a ‘proximate’, ‘evolutionary’, ‘ontogenetic’ or ‘historical’ point of view, so ‘cause’ (and ‘consequence’) might also be defined at different scales of analysis. What fraction in the network of interacting causal factors is truly accessible in science practice? For instance, the size of a bird apparently results from the accumulation of

11

numerous developmental processes between the onset of fertilization and the growth period that follows conception. Although all the details of these causal processes will not be accessible in science practice, impacts of a few external factors, like food consumption, may leave traces in expression of final body size. The quantitative link between input (e.g. food) and output (e.g. final body size) can only be determined with precision when the internal causal processes that transform food into biological expression are known. The identification of one or more underlying causal mechanisms may ultimately help to predict the future with different degrees of precision. Scientists often hypothesize that feeding nestlings will speed up growth. Nestlings that receive more food will grow faster. However, the exact shape of the growth curve requires knowledge of the network of the numerous biotic or abiotic causal factors that speed up or slow down growth. Growth will finally result from a continuous chain reaction in a complex and dynamic network of biotic and abiotic factors that continuously express spatiotemporal change and expression. Scale-dependence of physical expression and emergence Some philosophers argue that multi-component phenomena, like living beings, are not more than the sum of their components. Specialists in Physics following Lewes might claim however that specialists of bosons or fermions will not be able to predict at least some aspects in higher levels of structural organization because of emerging principles creating unique phenomena at different scales of analysis or perception. Particles or energy waves 'invisible' to the human eye may become 'visible' at certain levels of physical organization, perhaps because of new perceived functional properties not expressed at lower levels of physical organization. If billions of particles or energy waves are not clustered together they are not perceived by a human observer. However, when the same particles or energy waves cluster together they may become 'visible' to the same human observer and acquire newly perceived functions. Thus, some people would claim that a dynamic cloud of doves would be more than the sum of the individual doves composing the cloud. ‘Feasible’ projects Professional scientists from different research disciplines will most probably provide different responses to the same simple question from one unfamiliar field. As an example, hole-nesting passerines like blue tits (Cyanistes caeruleus) belong to the best investigated wildlife model species. Twenty children in a class room will definitely give different answers to the simple question how many eggs blue tits lay. If the same question is addressed to 20 professional scientists not familiar with blue tits, will the professional scientists score better than the children? Such a simple exercise may reveal that knowledge acquisition or science competence has its limits even for those that are top specialists in their own research field. The number of questions or problems addressed is obviously substantially larger in an education framework than in a feasible research framework. Feasible projects should define with precision logistic or resource constraints. The breeding programs conducted to understand genetics in Drosophila kept in test tubes obviously cannot be applied to understand genetics in non-domesticated avian wildlife that refuse to breed in captivity. A specialist in Avian Genetics could indicate this in a proposal. The definition of feasible research projects should indicate practical limits in individual research activity to avoid that potential believers, like decision-makers judging research projects, would force scientists to tackle problems that are perceived as ‘exciting’ or ‘urgent’, but that are not reasonable or unfeasible from a practical or logistic point of view.

12

Familiarity with science topics Science projects should be perceived as ‘feasible’ in at least one human-defined scale of analysis. In feasible projects, different contributors that work on the same scientific question should ultimately reach DSC when scales of analysis or perception are adequately defined. Proposing a scientifically feasible project requires time and at least some familiarity with model species, systems or methods. Proposing an idea to be tested may take five seconds. However, proposing a scientifically sound idea may take years when experience with model systems or literature helps to define an appropriate scale of analysis. Moreover, testing an idea may take years or decades if experience with model systems, literature and reliable data are required. The mismatch between the number of hypotheses proposed on one hand and the number of hypotheses truly empirically tested on the other hand is continuously growing. Hypotheses that were never scientifically tested are continuously accumulating in research reports or publications. ‘Optimality’ in curiosity How often do teams start to work with biological model systems without knowing the fundamental biology of species concerned or without directly consulting experienced people? To understand the functioning and evolution of a local model system with unique spatiotemporal characteristics, observation and experience will ultimately be more important than the knowledge from libraries dealing with model systems that express other characteristics. It is therefore unlikely that laboratory directors overloaded with administration tasks, or any citizen without adequate background knowledge, can identify a feasible science topic amongst a multitude of options defined by experts. Citizens might base reasoning on one observation (an anecdote) and think this may apply to any situation. But is this true? A ‘specialist’ that explores one flower during a whole life-time does she knows more about the true nature of the plant species than a ‘generalist’ that rapidly switches from one flower to another to explore amongst-flower variation in the same plant species? The definition of an expert evidently depends on scales of analysis or perception. An expert in Landscape Ecology has to cope with many environmental components and therefore will obviously be less detailed in the study of landscape component ‘A’ than an expert only dealing with component ‘A’ ignoring the other landscape components. Perhaps there is an ‘optimal’ level of curiosity accepting that costs and benefits of curiosity change when scales of analysis are changed. Individual specialization in one research field or one model system evidently will reduce competence at other scales of analysis. Science specializations might be defined as scientific casts. How to integrate knowledge from different scientific casts that work on the same scientific problem? If the ultimate goal is to better understand the true nature of nature, a natural phenomenon should be placed at the centre of investigation, not the scientific cast. Consequently, teams or laboratories should be organized to favor exchange between different scientific casts that work on the same problem. Medical sciences most often focus transdisciplinary research on a single body part in a single species. The specialized approach conducted in Medicine could perhaps also be advised in studies of other model species or systems. Ultimately, it is the scientific community as a whole that will provide solutions to problems and determine how the organization of the world is perceived, not the overambitious evidently biased individual researcher that wishes to master all knowledge from all scientific casts. If scientists cannot solve a scientific problem in a well-studied species can scientists solve the same problem in a less-studied species? Biology of humans and non-human vertebrates are likely at similar perceived levels of phenotypic and genetic complexity. Investigations indicate that non-domesticated birds and humans have many biological

13

characteristics in common. Humans and wildlife share many scientifically defined terminologies, including how organs, tissue, cells, or their phenotypic, physiological or chemical compounds are named. If success in research would be mainly influenced by financial support and human investment, humans that are studied in much more detail than other living beings may be considered as adequate models for the identification of knowledge frontiers. When a problem cannot be solved in human research it is unlikely that the same problem can be solved in research that involves less-studied animals. However, much science practice is not accessible to human research. Research in wildlife may reveal phenomena not investigated in humans because what is ethically allowed in research on wildlife is not (always) ethically allowed in research on humans. Alternatively, some research conducted with humans cannot be done with non-human animals. People may ask human creators why they developed or selected new human-made toys or why they artificially selected living beings, but they cannot ask wildlife why new mutations or new phenotypes appeared in nature. Predictive power in scientific research will depend on scales of analysis or perception, researchers’ experience with existing knowledge or logistic constraints. Citizens can predict future day-night cycles based on past empirical observation and scientists provide an explanation why day-night cycles will exist in the future. Climate or human economic dynamics are studied in much more detail than other environment-related dynamic processes. Such studies may indicate ability or inability to predict future environmental dynamics in less studied model systems. Scientists will be able to predict phenotypes of future clones at least at one human defined scale of analysis or perception when the techniques of cloning are sufficiently mastered and applied to different model species. Acceptable sample sizes that help to predict the future are definitely model-system dependent. Science terminology and feasibility Perceived feasibility of research topics may depend on how, and how detailed, phenomena are scientifically defined and described. This could be illustrated with research on exchange of living material or energy amongst living beings. The definition of ‘parasitism’ implies that living beings named ‘parasites’ take energy for individual survival at costs of other living beings named ‘hosts’. The terminology ‘host’ might give the impression to non-specialists, like tourists, that parasites are always mentally invited to take energy, whereas parasitologists assume parasites always take without giving. Humans can give energy (e.g. blood) without knowing the true identity of receivers, thus without knowing details of costs and benefits involved. How to make practically a distinction between 'taking', 'giving' or ‘sharing’ in Parasitology or Sociology when mental states of actors are ignored? The problem with wildlife is that you cannot ask in human language why they behaved as they behaved. Can parasites be scientifically identified without having access to mental states of organisms that transfer energy to parasites? Some scientists claim that invertebrate parasites may become collaborators of vertebrates when they stimulate vertebrate immune systems that improve defence against future exposure to organisms, like viruses, fungi or bacteria. When ‘parasites’ would select more vigorous vertebrate offspring in future generations, at what scale should parasite-related costs be defined? Perhaps costs expressed at the individual level are beneficial at the population level or vice versa. Do scientists have sufficient background knowledge to decide whether energy transfer amongst organisms involve parasitism or other forms of biotic interaction? Do parasitologists often collaborate with psychologists that focus on mental states to make a scientific distinction between ‘giving’, ‘taking’ or ‘sharing’ in biotic interactions?

14

When a human disease is named ‘epidemic’ only after a given fraction of the population expresses symptoms or declare complaints (e.g. fever) in medical services, how can this definition of epidemic diseases been appropriately applied in wildlife studies of which sampling methods are biased. For instance, in contrast to human patients, wildlife does not complain in human language or does not go to doctors to reveal symptoms to be included in statistics. Science terminology: theory versus practice In theory, each phenomenon is described with one definition to be applied in all circumstances. In practice, the same phenomenon will be represented by an unlimited number of biased measurements that always deviate from the ‘theoretical’ definition. For instance, the definition of 'body temperature' will theoretically be represented by one sentence or paragraph to be applied to any physical body that exists, biotic or not. By contrast, science terminology will practically be represented by a class of phenomena grouped together under the same heading ‘body temperature’. Body temperature has been measured using proxies, like skin temperature, ear temperature, or surface temperature. The terminology ‘body temperature’ will be presented in the titles or abstracts of scientific publications. How ‘body temperature’ was measured also revealing its deviations from theory will be found in the methods sections of publications. ‘Optimal’ science terminology In many research projects, phenomena are observed briefly, for instance to minimize impact of human presence following observation. This implies that science terminology based on brief observation or impressions have to accept potential imprecisions when phenomena are described in less detail. For instance, phenology of nest construction in small nest-box breeding passerines consist of several building stages, of which one is described as a 'pile of moss' expressed before the nest foundation is finished. The definition of this nest building stage is most often based on a visual impression to obtain a quick estimate of the breeding stage without counting or measuring moss fibres. A ‘pile of moss’ therefore represents a human-invented class that reflects numerous nest-stage expressions when details are taken into account. Does empirically measureable science terminology result from trade-offs between costs and benefits related to detail in measurement? Scientists may take the time to measure each detail of physical expression of each nesting stage. However, this may substantially increase time and energy-expenditure devoted to measurement perhaps also having scientifically ‘uncontrolled’ consequences. DSC is reflected in willingness of a science community to tell the same or similar stories based on arguments and perceived facts. Ability to reach DSC might be related to precisions required to define science terminology. Perhaps large-scale DSC is more rapidly reached when science terminology used requires less precision in science practice given that protocols may express significant spatiotemporal variation. Many teams may for instance agree that nest construction should not be measured in too much detail to avoid that measurement protocols have uncontrolled consequences for life-history stages expressed after the nest building stage. ‘Optimal’ science terminology that take costs and benefits of science measurement procedures into account would therefore obviously express spatiotemporal variation depending on which phenomenon is studied in the network of causally interconnected phenomena. ‘Optimal’ terminology as defined in science practice is expected to differ across research environments, even when scientists work on similar topics. Terminology based on ‘observation’ or the ‘interpretation of observation’?

15

When scientists describe (natural) phenomena they often replace the description of ‘what they see’ by interpretations of ‘what they think they see’, often without having sufficient access to background knowledge of how and why phenomena exist. Many introductions of study projects present ‘interpretations’ of phenomena instead of ‘descriptions’ of phenomena. These interpretations are often based on existing literature that describes and interpret ‘structurally or physically or behaviourally similar’ phenomena in other species or other study systems. Should terminology always be based on what is observed (‘Observation’) or what observers think they observe (‘Interpretation of observation’)? If terminology is based on what observers think they observe, the same phenomenon (e.g. observed behaviour: a bird approaching a feeder) might be defined in different ways because people differ in how phenomena are interpreted (e.g., interpretation of observed behaviour: when a bird approaches a feeder this might be interpreted as risk-taking, social behaviour, personality profile, exploration behaviour, hunger behaviour, etc.), perhaps also because scientists differ in biased education background (e.g. social versus physiological sciences). Critical judgements What is the difference between a critical remark and a scientific explanation for an unexpected result in a field experiment? Many field protocols have been developed because of published advice provided by similar studies tackling similar problems. If results of new studies obtain unexpected results, is it because of short-comings in experimental design or because local systems have unique biology-based or environmental-based characteristics requiring unique protocols based on unique local experiences. Example: If feeders are placed inside nest-boxes to improve rearing conditions of nestling birds following advice of published studies in other geographic regions, perhaps local characteristics including unidentified population-specific characteristics may result in unexpected results based on what has been published previously. Can in these conditions protocols be truly replicated from a methodological point of view?

Acquisition of data Hypotheses are empirically tested with observations or perception of phenomena involving more or less complex methodology. Observations are transformed into images, senses, symbols, words or numbers, named data. How observations are transformed into scientifically exploitable data differ between individual observers and environments of observation. Observers could biologically be defined as multi-component phenotypic structures of which different individual-specific components will contribute to what individuals will perceive and measure. There is not necessarily a close relationship between literature knowledge that results from language skills, field experience linked to physical performance, perception ability caused by eye characteristics and finger coordination ability that influences how scientific material is manipulated. Because of the unique biological composition of each individual living being, different observers do not perceive and measure the same environment in exactly the same way. Even for simple measures observer effects exist. A full description of a photo from a garden will definitely differ at least in some detail between two observers. The same observer might provide different descriptions during different time periods, perhaps depending on dynamics in mental focus or other biological aspects related to seasonality. Overlap in description across observers will definitely be influenced by the structural complexity of the phenomena described, the time devoted to observation and

16

language skills. Bias in monitoring might be reduced when more observers are involved. Observers might be considered as treatment or factor in empirical studies. Minor or unbiased observation errors do not necessarily change scientific conclusion. Interactions between observation and thinking: consequences for empirical research? Some people claim they cannot stop thinking. Does this imply they also think at the time they observe? Do people ignore phenomena when they simultaneously watch and think? If thinking would interact with observation, what are the potential consequences for empirical research? Do observer effects result from differences in mental states across observers at the time of observation? Unreplicable observation in less controlled conditions Networks involving citizens quantify unreplicable observations in scientifically less controlled conditions. This may happen when >1000 people quantify during 30 minutes per year different species of butterfly in private gardens. The species noticed might be rare or frequently observed. Only one reliable observation is required to consider the identification and presence of a species as true, but then how to know which observer is reliable? Should judgement of observation reliability depend on types or details of phenomena observed or techniques of observation used? Anybody can take a photo of a butterfly, but then the person making the picture has to be on the right spot at the right time. Observation reliability may differ between phenomena known versus unknown by science after controlling for observation protocol or observer experience. If phenomena express spatiotemporal variation, how many observers should focus on one study site and one time-period to avoid observation bias? If an international team publishes that the Amazonian forest harbours ca. 390 billions of trees who can and will verify this is indeed the case? Statisticians would predict variation in estimates of the size of the Amazonian forest. If another large-scale research network would replicate the study, and that network would provide another estimate, is it caused by dynamics of the Amazonian forest or biased sampling? Systems briefly visited Many model species or systems are briefly visited for scientific observation, because of logistic constraints or to minimize perturbations due to observation. Consequently, citizens that rear and live together with animals discover behaviour not observed by scientists briefly tracking individual wildlife from a distance. The details of bird behaviour in captivity will not be reflected in movement of a distant flying spot tracked with telescope. Pet animals may express less biased behavioural aspects related to intraspecific communication without human-induced stress components expressed in wildlife tracked by scientists. On the other hand, stress components expressed in wildlife might be considered as more ‘natural’. Perhaps scientists not familiar with model systems cause unnoticed or uncontrolled behavioural change in tracked individuals. Characteristics of phenomena might thus change because observers are present. Interestingly, novel findings are reported after more than half a century of research in the best-investigated wildlife model species. Does it reveal constraints in ability to observe wildlife or is it caused by temporal shifts in research interests? ‘Repeatability’ in field measures and observer effects Repeatability could be defined as the proportion of variance in a measured character that occurs amongst rather than within individuals. Estimating repeatability of measures or observations can examine reliability or consistency of measurements within individual observers for data sets produced by different observers.

17

Underlying causes of spatiotemporal variation in repeatability of measurements are either imprecisions caused by methods or scientifically relevant variation. Trained observers are considered more repeatable or consistent in measurement than untrained observers. Talented, but also 'untalented', observers might be revealed in scientifically-defined observation tests. Less talented observers can improve skills with the degree of exposure to phenomena. The interpretation or description of a phenomenon will ultimately depend on biological constraints that influence individual perception whatever the techniques used. Biology-based constraints that influence individual-specific perception evidently differ across observers. If one individual repeats the same behavioural act, detailed patterns of the same behaviour from the same actor will probably always differ. This can be illustrated with subtle differences in two signatures from the same individual, whatever the level of training. Biology-based constraints most probably influence repeatability of practical execution of science activity involving even more complicated behavioural acts in even more complex environmental conditions. Is mental focus or mental distraction related to ability to execute replicable protocols used to get data? Perhaps fatigue or individual biorhythms influence mental focus, and thus observation ability, at different moments of the day. How many people will notice environmental change at the time observations take place? Do students or scientists stop field observations when a car is passing by or an airplane is flying over or will this depend on which hypothesis is tested? Are observers more alert in the morning than in the afternoon or more alert before versus after a meal? Measurability causing observer effects may be related to what is measured and where measurements take place. More anxious target species move more when handled and are therefore more difficult to measure or track. Habitats exposed to harsher conditions may complicate application of standardised protocols. Some experiments may increase risk of measurement error lowering repeatability. For instance, tarso-metatarsi are more difficult to measure in smaller passerine nestlings. Thus repeatability in measures of tarso-metatarsi from experimentally enlarger broods reducing growth might be lower. Perhaps repeatability of phenotypic traits is related to environmental dynamics in food availability, food digestibility, parasitism, weather conditions or conspecific interactions. Repeatability in mammalian body colours may vary more or less with hormone state, depending on the temporal scale of analysis considered. Lack of replication in observation methods exist across empirical studies conducted by different teams that focus on the same topic. Dynamics in composition of local teams reflect dynamics in availability of trained observers per team. Teams also might differ in logistic constraints reflected in presence or absence of adequate methods to measure local phenomena. How can biology-based constraints and their link with uncontrolled observer effects be identified in a scientific research framework to improve data interpretation? Evidently, scientists are interested in biology-based causes of lack of replication in data not resulting from by-products of spatiotemporal variation in methods applied. However, philosophers might argue that humans and their methods are true products of nature. The mismatch between a phenomenon and the measurement of that phenomenon for at least one scale of analysis could therefore be considered as a natural phenomenon deserving more scientific study. Such studies can identify underlying mechanisms of human observer effects. Scientists sometimes mentally create data to test theoretical models using simulations. How to identify philosophically or scientifically mental observer effects in mentally created data sets?

18

Daily experience as data Theory is tested with empirical observation, but how often does observation during daily life provide inspiration in theory development? How long would it have taken to develop 'relativity theory' in Physics or 'parent-offspring conflict theory' in Evolutionary Biology without daily experience that inspires? Does theory result from empirical observation or vice versa?

Experiments If phenomenon ‘A’ causally determines expression of phenomenon ‘B’, a human intervention changing ‘A’ should temporally be followed by a change in ‘B’. ‘Experiments’ with scientifically acceptable design empirically test hypotheses aimed to identify underlying causal mechanisms of phenomena or relationships amongst phenomena. One factor called treatment can experimentally be changed to quantify the impact on phenomenon ‘X’ controlling for other potential factors that also may influence phenomenon ‘X’. Treatments, factors or confounding variables involved will change across model systems, species and hypotheses tested. When citizen actions produce a science topic Any experience from daily life could potentially be exploited to develop a scientific topic. As an example, citizens might put a chicken egg in a freezer without having any idea in mind. The initial observation is that the chicken egg freezes. A hammer will break an egg at room temperature, but will not break the egg when exposed for one day at -17°C. The observer may think there is a causal relationship between the state of the egg and the ambient temperature, independent from other environmental factors. How will this observation be further exploited? Perhaps the observer will think egg freezing is so obvious that it is not interesting enough. The observation might also be perceived as unexpected stimulating curiosity. Someone else hears about it and may get ideas about consequences of egg freezing. Freezing probably sterilizes eggs. Did birds evolve strategies to avoid egg freezing or to anticipate environmental conditions that avoid egg freezing? Are temperature-thresholds that result in egg freezing related to egg characteristics? Smaller eggs will probably freeze more rapidly than bigger eggs. Perhaps arctic parents include antifreeze chemicals into eggs. Do eggs from tropical birds freeze faster than eggs from arctic birds after controlling for egg size? Perhaps egg layers anticipate favourable periods, such as periods with higher ambient temperatures, or avoid unfavourable periods, such as those including freezing temperatures. If a chicken egg is placed in a winter glove in a freezer, egg freezing will probably be delayed. Perhaps birds evolved protection strategies to retard egg freezing. Egg freezing could perhaps also be delayed when females roost on eggs. Do birds add more lining to cover eggs in colder regions or during colder years? Do arctic birds use more lining to protect eggs against extreme weather conditions compared to tropical birds? Both citizens and scientists have access to meteorological data about probabilities of freezing at different places in the world. This information may be exploited to predict when, where and how birds will build nests and lay eggs. Thus, initial ideas for the development of an interesting scientific topic can be obtained at any time and at any place. Other experiments conducted at home Any person been able to install an aquarium or terrarium to observe fish, reptiles or invertebrates can use the tool to investigate behaviour in more or less controlled environmental conditions. For instance, Paracheirodon sp. can be bought in pet stores by any

19

citizen not possessing certificates approved by ethical or veterinarian commissions. Citizens at home may observe that claimed ownership is dynamic and rapidly adjusted to changes in the local environment. They also may observe that communities composed of species with different behavioural activity patterns or personality profiles will influence where individual species will live and reproduce. Phenomena investigated by professional scientists could thus also be investigated by amateurs, also to improve living conditions in captives. Perhaps scientists discover aspects about species known for years by amateurs. Such experiments not only can be used to teach Behavioural Ecology, but also to investigate biology-based constraints in science practice, like underlying mechanisms causing observer effects. Experimental design and constraints in execution Review papers often high-light shortcomings in experimental design. However, there will always be at least one inaccessible scale of analysis or perception in laboratory experiments. Laboratory conditions will never be truly standardized from a practical or logistic point of view. A test cage or a laboratory harbouring a test cage consists of multiple visible and invisible characteristics reflected in methods to clean cage environments, individual observer behaviour, material coming from multiple ‘uncontrolled’ external sources, etc. Field conditions may scientifically never be truly standardized because of perception constraints, logistic constraints or continuous dynamics in nature. The need to standardize experimental environments may depend on the scientific problems addressed or model systems considered. As an example, between-population variation in timing of reproduction in blue tits is sometimes larger in scientifically more standardized aviary conditions than in scientifically less standardized free-ranging conditions. Responses of blue tits to ‘artificial’ versus ‘semi-artificial’ conditions may result from scientifically ‘uncontrolled’ organismenvironment interactions. Not all study populations may be pre-adapted to captivity and scientists most often don’t know this when they initiate laboratory experiments. Veterinarians may advise to use sterilized test cages minimising infection. The practical problem is that blue tits will not breed in these highly artificialized environments or that some populations cope better with captivity than other populations. Outdoor aviaries with ‘semi-controlled’ natural vegetation might be more appropriate for captive wildlife breeders, simply because they better simulate wildlife conditions. How can a veterinarian or members of ethical commissions that never worked with blue tits provide constructive advice about both scientifically and ethically acceptable experiments? Should ‘environmental sterility’ or the ‘mental state of captives’ been used to make decisions about how to conduct experiments? Do sanitary recommendations proposed by people not familiar with model species complicate execution of scientifically acceptable experiments that take background knowledge from model species into account? Sanitary conditions are never truly controlled in wildlife conditions anyway. Lack of replication in design or execution of experiments may be attributed to unique spatiotemporal characteristics of study plots or other factors expressed at the time field experiments are conducted. Study plots for long-term field observation are used to experimentally test hypotheses of which biological effects can leave traces in following generations of model species. How to identify the relative importance of human versus nonhuman induced effects in study plots used for long-term investigation of wildlife? Details of field experiments will never be fully replicable across years. Plants and trees in study plots grow. Study populations and research teams investigating them change in composition. Detailed environmental conditions for field experimentation therefore also change.

20

Storing, maintaining and sharing data Samples or pictures from nature or observations transformed into data are stored in note books, boxes or electronic files that have to be maintained for subsequent analyses. There might be a mismatch between details in contents of note books reporting field or laboratory observations and data subsequently electronically summarised and stored for analysis. Human-induced errors during data transfer from one medium to another cannot always be fully excluded. Biology-based mental focus during data transfer evidently expresses spatiotemporal variation. Perhaps there is a relationship between the complexity or size of data sets and the probability to introduce copy errors. Some scientists relying on probability laws claim that data sets without human ‘errors’ do not exist. Data managers differ in efficiency in how samples or data are stored and maintained because of different underlying mechanisms. One person is most often locally in charge to verify data stored in local computer files, but data managers do not always have experience with field or laboratory work. Data might represent several hundreds of data lines per study site per year. Macro-geographic studies often require fewer data lines per study plot than micro-geographic studies often working at much finer scales of analysis. Are data file verifications more or less efficient when more people are involved? How important are 5-8 data lines in a file with 500 data lines? How important are these 5-8 data lines when they represent 30 years of research and education activity? (Re-) verifying data sets are not a waste of time, also from an ethical and human social point of view. Data reliability is evidently a priority, not the speed or ease of data sampling or storage, and not the methods used to store or maintain data. Simple note books stored on a shelf in a library might be more efficient to safeguard original data than complex sophisticated computer systems that cause unnoticed change in composition or organisation in electronic data sets. Perhaps electronic long-term data-bases suffer from human-induced computer crashes or electronic viruses. Perhaps dynamic computer systems that provoke unnoticed change in data sets resemble the biological organisms that are exposed to unnoticed genetic mutation. What are social, scientific or financial consequences of uncontrolled change in data ignored by people that store or maintain data? Principles in communication indicate that messages between senders and receivers are transformed during transfer. The speed of transfer or degree of transformation might be environment-dependent. Features of biology-based communication involving multiple senders and receivers probably also apply to interconnected human-created computer systems. Research based on collaboration networks evidently exchange raw data between local or distant collaborators via normal post services or electronic computer services. But how to verify that data files have not been transformed during transmission between senders and receivers? Data may also be lost when computer systems are abandoned and staff members retire. Laboratory teams or factories that developed complex techniques must remain available for assistance, maintenance, correction or improvement. Re-use of information How much data remain hidden in note books or study reports because students or staff members leave a laboratory? How to recycle or reuse (‘old’) data in the absence of people that collected them? Perhaps those that exploit long-term data from others do not have access to details related to monitoring protocols or experimental design. Who should store information concerning ancient experimental protocols that produced data to be re-exploited in new science frameworks: the scientist, the student or technical staff? When old data or samples do not have scientific value anymore? For instance, when 8 million plant samples are stored in a museum, what fraction will be exploited or exploited more than once in a science framework?

21

Maintenance costs of samples stored in refrigerators for decades and thrown away without exploitation might be substantial.

Analysis of data Statistics provides man-invented rules for data analysis. Statistics could be defined as a tool to describe patterns, like functions not indicating causality. Underlying causes of the patterns observed can subsequently be discussed. Bias in statistically created patterns might favour or exclude explanations that data analysers had in mind. Data can visually be analysed by inspecting tables or figures giving initial ideas for analysis. Conclusions based on visual inspection of figures or tables might differ from conclusions based on output of analyses from statistical packages simultaneously taking many variables and factors into account. Data sets from well-planned experiments in well-studied model species probably require simpler analyses facilitating data interpretation. Using very complex statistical analyses might impress citizens or students, but they do not necessarily improve confidence in conclusions. Accepting that whole research careers are devoted to Statistics alone, and journals only devoted to Statistics continuously produce new publications, how can a student, scientist or teacher from any research domain become expert in statistical analysis? If 20 data analysers with different scientific backgrounds would be asked to analyse the same data set, details in statistical procedures will probably differ. Perhaps overlap in interpretation of results will depend on the complexity of data sets and statistical procedures carried out. Researchers use different statistical approaches to study the same scientific problem with the same model species. This may reduce replication in statistical procedures across studies as revealed in methods sections of scientific publications. Researchers have the option to blindly follow advice from statistical text books how to transform and analyse data. But why should raw data with skewed distributions be less informative than transformed data that impose ‘normality’ or other data distribution types? Why do statisticians impose data transformations (often) without biology-based foundation? Perhaps published significant statistical effects will simply depend on data treatment, such as transformations that improve ‘normality’ to increase the probability to find statistically significant results. Are initial analyses mainly based on trial and error aimed to guide data analysers in more detailed exploration of the same data set until statistically significant, and publishable, patterns are found? Perhaps strong statistical effects are insensitive to the types of statistical tools or data transformation procedures carried out. Each research laboratory might develop or possess a platform involving statisticians assisting specialists from other research fields in data analysis. This is done in the pharmaceutical private sector where specialists in Statistics statistically verify biological impacts of new molecules in tissue or model species using data provided by experimental biologists. If specialists of model species do not believe the results from statisticians that worked with data provided by specialists, there might be an unidentified problem with experimental design or the types of statistical procedures carried out. Statisticians should ideally interact with those that are familiar with model species or systems to ultimately provide repeatable and biologically-relevant findings. These interactions should ideally be initiated prior to the onset of empirical studies.

22

‘Unreplicable’ research findings might be attributed to different causes. Lack of communication amongst teams prevents perceived replication of protocols or experimental design. Samples from different places might have unique characteristics. Data analysers also have unique education backgrounds perhaps having more or less experience with a statistical tool or model system. It is therefore not a surprise that published, and thus scientifically accepted, statistical analyses are rarely perceived as truly replicated across studies. Moreover, how many unpublished statistical tests will be applied before a final test will be accepted as the best one to be published? Are preliminary statistical tests truly replicated across studies that involve different teams? Philosophically, details of statistical patterns will never be replicable whatever the experimental design or standardization procedures applied. There will always be at least one scale of detail causing two statistical patterns to differ. The true philosophical challenge therefore might be to explain why two statistical patterns differ, not to explain why they do not differ. Limits in human-invented mathematics to capture nature’s complexity Mathematics and Statistics are human inventions aimed to describe nature’s diversity and dynamics. As an example, numbers could be considered as human-invented symbols aimed to count and quantify phenomena (objects, events, organisms) that have human-defined characteristics in common. Mathematicians add physical phenomena with perceived shared characteristics (1 + 1 = 2). However, two physical phenomena always differ in at least one level of analysis or perception (1' representing phenomenon A does not equal 1'' representing phenomenon B) and phenomena differing in Physics are given the same mathematical weight. Consequently, 1' + 1'' does not equal 2, or 1 + 1 = 2 does not reflect Physics in nature. The need to count with numbers will thus depend on physical details taken into account when phenomena are defined or described. If each individual is unique in physical structure, quantities exceeding '1' are philosophically not required. In other words, the simplest mathematical based summary of nature's complexity would be '1', and the types of '1' to be mathematically defined would match the number of physical structures investigated. Can statisticians with highly simplified visions of the world impose analyses to biologists accepting nature’s true complexity and diversity? Scientific observation is translated into forms that can be statistically analysed. Statistical analysis requires that some variables are presented as ‘continuous’, others as ‘classes’. When carbon (‘C’) is used to define a statistical class, all living beings or objects containing carbon might mathematically be grouped together to define a class with ‘C’ versus a class without ‘C’. However, the potential error or imprecision with this classification is that each individual ‘C’ atom is probably unique in physical expression when all empirically accessible or inaccessible scales of analysis are included. The mathematical definition of a class should therefore be considered as a relative and simplified concept accepting ‘imprecisions’, simply depending on background knowledge or perception ability of those that define a class of phenomena having perceived characteristics in common. How do Psychology studies classify human emotions from a statistical point of view? Data analysers have to believe what people tell how they feel and try to translate what people tell into analytical classes having at least some described within-class features in common. This does not exclude that different people having the ‘same’ perceived feeling provide different descriptions perhaps also depending on language skills. Different analysers might interpret the same description in different ways perhaps also depending on which aspect of the description (e.g. one word or one sentence) is individually perceived as more or less relevant. Thus, when mathematicians or statisticians propose models for the functioning of natural

23

processes, do they have sufficient background knowledge in Ecology, Biology or Physics to capture nature’s complexity?

Relativity in statistically defined parameters: the case of data distributions The definition of data distributions may depend on background knowledge available to analysers of data. For instance, an ornithologist observes that clutch size in a tropical avian model species varies between 4 and 8 eggs and obtains a bell-shaped frequency distribution when the scale of the X-axis is defined as 4, 5, 6, 7 and 8. However, a fish specialist may not use the same arguments for the definition of the X-axis used to present the distribution of the avian clutch sizes. Fish females usually lay between 10 and >1000 eggs. A fish biologist unfamiliar with values of clutch sizes in tropical birds might justify the definition of an X-axis that pools together clutch sizes varying between 1-10, 11-20, 21-30, etc. based on data from fish, poultry science or private observations. Domesticated chicken at home continuously produce eggs every 2-3 days, except during short-day winter periods. Why not defining an avian clutch size axis with values exceeding 100 eggs? Perhaps the fish biologist would have placed the clutch size data from the tropical bird in a single class (class 1-10) with an X-axis providing clutch size values between 0 and >100 eggs per clutch. Thus data distributions and subsequent statistical analyses of the same data set might vary depending on baseline knowledge available. Data analysers inspecting tables or figures may decide to exclude from data sets unusual data points named ‘outliers’. The outcome of statistical analyses will probably differ in analyses with versus without outlier data points. The identification of outliers may depend on statistical rules taking observed variation into account (e.g. points exceeding or not exceeding standard deviation values, which may be sample-dependent) or familiarity with model species or model systems (e.g. data points considered to be biologically impossible, perhaps caused by copy errors). However, underlying mechanisms creating outlier data points are usually unknown. People might always find published or unpublished arguments to exclude or keep ‘unusual’ data points in analyses. Perhaps selection of data points will depend on baseline knowledge of data analysers and, why not, personality profiles, like more or less critical people independent from education background. Time-lag between availability of new tools and its wide-spread application Computer business and development indicate that replacement of old computers by new ones is slower for individual byers than for a community of people permanently offering new computer devices. The same reasoning could be applied to any science device or tool of which techniques evolve fast. How many published studies ignore existence of new tools used for data analysis? If a new (e.g. statistical) tool is proposed in a publication how long does it take before the tool will be applied by a large fraction of the science community concerned? New propositions for data analysis are numerous and evolve fast. The time-lag between the proposition of two new tools might be substantially shorter than the time-lag between the proposition of a new tool and its wide spread use. Does it imply that the vast majority of the studies do not use the most advanced tools for data analysis? Given the fast technical developments, how long should people wait to select or learn an appropriate tool? If catching up with learning new tools is a fulltime engineers job, and specific tools are rarely used by the vast majority of the scientists concerned (e.g. biologists, ecologists), a better option might be to be assisted by specialists that track and apply the most advanced techniques. This is probably more efficient than that scientific generalists

24

continuously learn older techniques in a technical world that evolves (much too) fast. What are costs and benefits of multiple laboratories providing exactly the same technical services, e.g. used for molecular, physiological or other measurements? How to coordinate more efficiently technical-assisted collaboration at national or international levels, for instance to avoid individual scientists have to learn everything what is available?

Interactions between methods and hypothesis-testing There exists an uncountable number of ways to gather and exploit information. People might think that the more information they have the more they will learn how the world functions and evolves. People might cumulate observations without having any pronounced idea or hypothesis in mind. However, scientists have to make choices amongst the uncountable number of options for observation and analysis available. Biased data sampling is therefore usually guided by ideas or hypotheses scientists have in mind. Data sampling and analysis procedures evidently influence how hypotheses will be scientifically tested also influencing data interpretation ability. Findings from one analysis might result in one or more conclusions rejecting or accepting one or more hypotheses, most probably also depending on biologybased background knowledge of observers involved. Impact of methods used for data sampling to test hypotheses, and impact of detail in analysis on ability to test hypotheses, is illustrated with following hypothetical examples. Analysis of short wildlife films A film lasting about four minutes on YouTube might illustrate examples of wildlife behaviour, like plant-animal interactions during pollination. The film can practically be subdivided into a large number of pictures on a screen of 1 m². Each individual picture of the film might also be subdivided into smaller or larger compartments. If one picture would be subdivided into a puzzle of 10.000 cm² and each 1 cm² is visualized and analysed separately, which part of the screen is required to get an idea about the information provided by the whole screen and the whole film? Doing this might result in a continuous flow of an uncountable number of potential hypotheses. Different people watching the whole film, a fraction of the film, or a fraction of the screen presenting one picture of the film will produce different hypotheses based on what they perceive and what they memorized in the past. Different potential hypotheses will definitely follow different probability distributions depending on who is watching the film and scales of analysis or perception involved. The probability that two observers will produce exactly the same hypothesis using the same wording and terminology will probably be very low. How many hypotheses might be formulated focusing on only 1 cm² from a screen of 1 m² and what will be the scientific approach used? Hypotheses and methods used to test them will definitely change with changes in scales of analysis and perception, e.g. going from 1 pixel to 1 cm² to 2 cm² up to 1m². Pixel analysis might focus on physics of colours whereas analyses of whole pictures might focus on physics of colours or colour contrasts but also on other aspects, like how and why two filmed species interact. Moreover, how many mathematical equations will be required to describe film details concerned with pollination of flowers? If people only would have access to Mathematics describing the film they did not see before, how would they translate equations into film pictures? How would different mathematicians cope with 20 pages of mathematical equations to be transformed into visual pictures of which contents become accessible to citizens? All this illustrates that interactions between methods and hypothesis-testing involving people varying in background knowledge might potentially result in an uncountable number of hypotheses and related conclusions, evidently depending on

25

scales of analysis or perception involved. It also implies that the same problem might be tackled in many different ways depending on who will start and conduct research, probably related to past biology-based experience of the researchers involved. How to quantify motivations underlying behavioural expression? The interpretation of wildlife behaviour from a distance is not easy. Scientists cannot ask wildlife to explain in human language why they behaved as they behaved. Motivation expressed in the mind will influence behavioural action, but the same perceived behaviour might result from different underlying motivations or feelings. This can be illustrated with scientific observations from a distance that one individual only drinks wine whereas another individual only drinks water. How to interpret this observation? Specialists in taste science might propose that the two individuals drink what they like to drink: one person likes wine, whereas the other person likes water or does not like wine. Specialists in social sciences may argue that the two individuals like wine, but that one of the two drinks water because the other prefers wine. Health specialists could perhaps propose that the individual that drinks water only drinks water because others told him that too much wine penalizes personal health condition. Wildlife specialists might explain that the person only drinking water does not want to share the bottle of water with the other person. Specialists in religion might think that the person that drinks wine is forcing the companion to drink water so that the companion remains healthy. Specialists in natural history might want to know the role of historical facts. Did the person that drink water drunk too much wine the day before? How can observers analyse and interpret wildlife behaviour from a distance without having access to information provided by individuals explaining why they shared or not shared resources or without having access to historical facts prior to the initiation of observations? Apparently, there are many possible interpretations based on a few simple observations in the field or a cafeteria. Ideally, different hypotheses should propose different predictions perhaps requiring different tools or methods to test them. For instance, consumption constraints could be decreased by giving two individuals >2 bottles of wine and >2 bottles of water. The order of consumption of all the bottles received might test the ‘taste’ hypothesis in more detail or could verify if two individuals exchange bottles differing in content, and if so, in what direction. Blood tests or other physiological sampling methods might indicate overconsumption of resources (e.g. wine) in the recent past or in the absence of observers. Wildlife psychologists or ethnologists might accept or reject some hypotheses through analysis of face expressions, gestures or social interactions. Biology-based mechanisms that influence perception and therefore data sampling Psychologists sometimes present paint spots to patients to determine how they respond. Different patients tell different stories when exposed to the same paint spot. Perhaps this might also be the case when different data samplers are confronted with the same photo, the same question, or the same publication. Perhaps judgement of photos, pictures, art objects or any natural phenomenon observed expresses spatiotemporal variation because of biologybased underlying mechanisms, like season or light-exposure related individual perception characteristics. Can we fully exclude that perception or attraction states are influenced by hormone or fertility states of data samplers? Do scientific sampling procedures take biological rules of mental perception and interpretation into account? Differences in composition of study and target populations Compositions of study populations should be representative for compositions of whole target populations. Critical text books will argue this is not always the case. How reliable are polls when no information is available about characteristics of individuals sampled? Perhaps people

26

that participate to polls differ from those that are not willing to participate thus providing biased pictures in opinion. If the majority of a sample of 1000 people in a poll states they keep supporting a system of BAC exams, is this because the vast majority of participants succeeded with BAC exams? If the same poll would be addressed to students 1 to 3 years before BAC exams, would the majority be against BAC exams? Perhaps people change opinion concerning usefulness of BAC exams when they know more about performance of education systems without BAC exams. Not all characteristics of individuals composing study populations are consistently sampled. Characteristics that require considerable financial or human investment in sampling include techniques involving molecular markers, physiological markers or behavioural aspects, like personality profiles. These characteristics are most often sampled for short periods of time in long-term field studies or sampled in a limited number of biased study populations. Perhaps population characteristics and their environments are better controlled in small-scale laboratory studies that use selected strains or breeds as biological models. Unicellular algae have been placed in containers to examine across generations clustering to study evolutionary transitions between unicellular and multicellular life. Highly controlled environments that are more efficiently monitored by scientists are evidently not always representative for what is happening in natura. Thus, observations using test tubes of which results do not have demonstrated high external validity might examine frontiers in humancreated theory or discover natural phenomena not reported in previous studies. Data sets, statistics and data interpretation All statistical patterns observed have underlying causes whatever the size of the samples analysed. A sample size of ‘1’, such as a newly discovery planet without an orbit around one sun, might provide clues about sampling sites or sample characteristics. All statistical patterns have underlying causes. Statistically non-significant results are definitely significant from a physical or biological point of view. The challenge will be to identify underlying causes of patterns, whatever the shape of the patterns observed. Should one only look at broad statistical patterns in data clouds to interpret biological data, or should one also focus on individual data points or ‘outliers’ to provide propositions for underlying mechanisms? In a visualized correlation that presents individual body size against individual body weight, different parts in the data cloud might result from different underlying mechanisms. How much biological or ontogenetic processes have been involved in the ultimate expression of individual body weight and individual body size at the time of measurement? Each individual data point is evidently caused by a unique individual-specific cocktail of underlying biological and environmental factors of which the details are most often empirically and logistically not accessible at the time of data analysis. Local environmental or biological factors are often so dynamic that proposition of hypotheses without access to background knowledge of study systems or species often makes scientifically no sense. For instance, biologically irrelevant timing of sampling in highly dynamic environments will often determine the outcome of statistical patterns and determine whether hypotheses will be rejected or not. If >80 specialists are allowed to share and analyse a data set and there is consensus concerning data organisation and final analysis among >95% of the contributors, it is unlikely that results and conclusions will change when additional people would verify and analyse these data. However, in a fast evolving world, it might be advisable to reanalyse the same, or an enlarged data set, with 10-year intervals taking new knowledge into account. New knowledge might allow the redefinition of factors included in data bases.

27

Research domains differ in acceptable empirical science procedures Scientists often make claims addressed to citizens without defining scales of analysis or perception. If astrophysicists claim that 80 specialists that used nesting boxes as tools. Clutch size is one of the best studied life-history traits in wildlife. Proximate and evolutionary mechanisms potentially influencing clutch size expression have been discussed by several generations of ornithologists during the last 60 years. The problem of causes of variation in clutch size might be perceived by citizens as a simple scientific question tackled with simple methods often also accessible to citizens. However, it took more than one year to reach consensus concerning major causes of clutch size variation in the best studied avian model species. Accepting that each of the >80 specialists would have decided to continue working and publishing without open-access exchange, how long would it have taken to reach DSC at the science community level? Another 60 years? Open access exchange of data, methods and ideas might be a solution to reach consensus more rapidly. Proposing solutions to solve problems more rapidly should be one of the major scientific tasks, or not always? What criteria are used to select collaborators for macro-geographic studies, besides science politics? Why are some countries missing in large-scale studies? What is the impact of biased selection of collaborators on research findings?

Concluding Remark Scientists might work hard, but efficiency will ultimately also depend on efforts from visible or invisible contributors in communication and interaction networks, dependent or independent from science activity.