Simon Reader:
(Thursday, June 28, 2pm)
Learning from others (social learning) is extremely widespread. Individuals from a variety of species acquire foraging, anti-predator, mating, or habitat information from the activities of other individuals. This social transmission can have significant consequences. For example, alarm or ‘fear’ responses to novel stimuli can be acquired and transmitted between individuals. What animals feel and know may thus be partly the product of what others feel and know. Cocial learning had been thought to require cognitive processes that evolve and develop independently, as adaptive specializations that optimize learning from others. More recently, much variation in social learning is thought to be the product of experience shaping general associative learning capacities, with evolutionary effects on social learning limited to changes in ‘input’ mechanisms rather than central cognitive processes. A similar debate concerns the origins of “mirror neurons.” I will discuss the evidence and implications of this shift in viewpoint about how animals understand and represent others.
Reader, S. M., Hager, Y., & Laland, K. N. (2011). The evolution of primate general and cultural intelligence. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 366(1567), 1017-1027.
Reader, S. M., Morand-Ferron, J., & Flynn, E. (2016). Animal and human innovation: novel problems and novel solutions. Trans Roy Soc B 20150182
Street, S. E., Navarrete, A. F., Reader, S. M., & Laland, K. N. (2017). Coevolution of cultural intelligence, extended life history, sociality, and brain size in primates. Proceedings of the National Academy of Sciences, 114(30), 7908-7914.
Reader, S. M., & Laland, K. N. (2002). Social intelligence, innovation, and enhanced brain size in primates. Proceedings of the National Academy of Sciences, 99(7), 4436-4441.
I am really not knowing much on social learning in fish behavior, but from what I learned listening to your presentation, I was actually wondering until what point we can call it « social learning » when an animal has and alarm or « fear » response to a new stimulus. I am thinking of the turtles. The female turtle puts her eggs in the sand and then, when the eggs hatch, all the little turtles got to get to the sea without being eaten by the predators. However, nobody or nothing is teaching that they got to go to the sea. Do you have any idea how do they know? It this behavior is as innate as breathing, for example, or you would say that it is a kind of social learning? Are the innate behaviors of species of animals, for example of the fishes, known? Because, I am wondering what if the fear response to a predator could not be innate instead of « acquired » by social learning. Therefore, I guess my question is essentially : innate or acquired?
ReplyDeleteForgot to say also that I know you explained us your research with guppies that followed and didn't, with or without food, but do you know if other researches were made and had the same results as you?
DeleteThank you for your questions. Yes, certainly, genetic predispositions can underlie alarm and many other behaviours. For example, there is evidence that fish anti-predator behaviour can be the result of evolved genetic predispositions, that it can be the result of experience (such as social or individual learning), and that it can be the result of the combined effects of genetic predispositions and experience (e.g. experience could have a greater effect in fish that have evolved in high predation rather than low predation conditions, as has been found in minnows).
DeleteAs a side-note, many ethologists (including myself) tend to avoid the term ‘innate’. One reason is that it has multiple meanings and one meaning (e.g. present at birth) does not necessarily imply another (e.g. unlearned) (Patrick Bateson and colleagues have written on this topic). Another reason is that ‘innate’ tends to suggest behaviour is the result of genes or environment rather than genes and environment.
This is the full citation for the bumblebee study I spoke about, that has similar results – bees can be trained to use or avoid social cues:
DeleteDawson, E.H., Avarguѐs-Weber, A., Chittka, L. & Leadbeater, E. (2013). Learning by observation emerges from simple associations in an insect model. Current Biology, 23: 727-730.
(You can get the PDF on Lars Chittka’s website – he is speaking later this week)
You warn us about adaptationist just-so-story and I think it is a very important consideration... I agree that this is not simple task to establish if social learning is an adaptation; not clear if it is an adaptive specialisation or a general process… But, what if innovation, associative and social learning just come from the same exaptative pool that Gould taught us about? I think variations and natural selection are the motor of evolution, but I think that biologicals constraints channel evolution. What if flexibility is the result of the gap left by these biologicals constraints? In this line of taught, studying these constraints could be insightful. What do you think about that?
ReplyDeleteThanks for your interesting question. Yes, I can certainly see exaptation being important in social learning. If social learning is the product of general learning processes, as has been argued, social learning would be an exaptive byproduct of the evolution of individual learning. One could imagine a situation where a tendency to use social learning emerges from developmental processes (e.g. because social learning is rewarded in that particular environment) but then evolution acts to increase the efficacy of social learning (e.g. perhaps perceptual mechanisms become tuned to social stimuli). I agree that examining constraints on social learning would be insightful.
DeleteThanks for your answer and your support.
Delete
ReplyDeleteI wanted to follow up on my question about the 'Killjoy' issue with a couple of comments. I've said this a couple times before and I don't intend to say that competing explanations are often or even usually demonstrably equivalent, but I do think that, a la Quine, scientific theories are underdetermined, and the way that we understand phenomena is determined by the paradigm in which we are operating. While your approach to explaining behaviour is valid in some purely empirical sense, I don't think that it necessarily rules out or should necessarily be preferable to other explanations which invoke an internal experience or some sort of 'planning.' While the principle of parsimony may be a useful scientific tool, I think it needs to be more carefully applied if you're extending it to a metaphysical or ethical domain of inquiry. While it may be efficient to have a parsimonious explanation of fish behaviour for scientific reasons, if this purely external behavioural model leaches over into the realm of ethics as the predominant paradigm, for example when deciding on a humane way to slaughter a fish, I think the consequences could be extremely negative.
Thanks for this follow-up. To be clear, I’m using ‘killjoy’ in a particular sense, perhaps with a different meaning (see below). I used the term as I wanted to be clear that I am sympathetic to simple explanations of animal behaviour, partly because I have experienced the assumption that work on animal innovation assumes complex explanations. My view is that we shouldn’t assume certain human-like mental processes (e.g. intentional deceit) to other animals without considering alternative explanations. I’m certainly not arguing that fish or other animals are automata – this isn’t at all what I meant by ‘killjoy’.
DeleteHere a quote from a recent symposium on the topic (the 2015 ISHPSSB, at UQAM too) that captures what I mean: “Comparative psychology has been disparagingly described by some as a field of “romantics” and “killjoys”; some researchers being keen to attribute human-like mental states and cognition to non-human animals, and others arguing that to do so is anthropocentrism.” Shettleworth provides a relevant discussion of the term (2010: ‘Clever animals and killjoy explanations in comparative psychology’, Trends in Cognitive Sciences).
I’m arguing that ‘killjoy’ explanations should be considered, and that they should be tested, not assumed. I was talking specifically about social learning, and so I am not saying that if the associative-learning explanation holds for certain types of social learning that it automatically holds for other realms of animal behaviour or even for all types of social learning– those are separate questions. From the ethical point of view, and specifically fish welfare, one of the most important questions is whether fish suffer. If anything, my feeling is that demonstrations of social learning in fish raises welfare and ethical concerns in fish rather than minimizing them. For example, if stress responses can be socially transmitted, then we have to consider both direct and indirect (social) effects on fish wellbeing. I also feel if killjoy explanations are not examined – for example, the necessary control conditions are not run in an experiment - there is a risk that the research is not widely accepted, and thus has a smaller impact on wider issues than it could otherwise have. So in short my message is that we should test, not assume, our proposed explanations for behaviour.
L’apprentissage social se passe chez plusieurs espèces animales. Actuellement, il semble y avoir des corrélations entre l’intelligence sociale et l’intelligence individuelle chez plusieurs espèces animales. Des espèces sociales, telle que les grands singes, les éléphants et les cétacés, sont de bons exemples d’animaux très intelligents et très sociaux. D’ailleurs, comme ils vivent en groupe, ils pourraient d’avantages bénéficier des effets de l’apprentissage social. Ces espèces ont aussi de très grands quotients d’encéphalisation, c’est-à-dire que la masse de leur cerveau est beaucoup plus grosse que ne le prédit la taille de leur corps. Il y a définitivement un facteur d’apprentissage social (ou individuel) associé à la taille du cerveau, même s’il n’est pas très claire en ce moment. Il est possible qu’entre des espèces phylogénétiquement proches, les cerveaux plus gros permettent des capacités cognitives supérieures. Par exemple, les humains ont les plus gros cerveaux parmi les primates. Sont-ils cognitivement supérieurs? Un point de vue très mitigé!
ReplyDeleteThank you for your comment. I think it is challenging to compare human cognition with the other great apes because culture has such a dominant role, although it is certainly interesting to do so. I think we also have to be careful to characterise species as ‘cognitively superior’. Animals may perform well on some measures of cognition and not others.
ReplyDeleteL’apprentissage social n’a pas seulement lieu dans le règne des animaux humains, mais également dans le règne animal non humain. Effectivement, il est extrêmement répandu. Il peut être rapide, s'accommoder aux circonstances, aux différents contextes, et altérer grandement le développement. La manière dont on perçoit et reconnait les autres est le résultat de processus sociaux. Cet apprentissage social est sans doute important dans la façon dont les différents animaux sociaux se comportent. Il se pourrait bien que la “théorie de l’esprit” que l’on a à propos des autres dans l’environnement soit issue de cet apprentissage.
ReplyDeleteJ’ai trouvé bien intéressant la question concernant les implications du type de processus (spécialisation adaptative vs processus général d’association) que représente l’évolution de l’apprentissage social par rapport à la représentation de l’autre. Il semble, comme le mentionne professeur Reader que l’adaptation spécialisée permettrait de voir les autres agents sociaux comme d’autres individus, participant possiblement à notre apprentissage plutôt que comme de simples signaux potentiels. À cet égard, l’expérience avec les poissons montrée au début de la présentation référerait, semble-t-il, plus au processus général d’association ; les poissons du bocal démonstrateurs agissent ici comme signal de la présence d’un danger que les poissons du bocal observateur associe à l’odeur initialement non aversive de la morpholine.
ReplyDeleteJ’ai cependant de la difficulté à me positionner face à cette distinction. Il semble effectivement que l’apprentissage social puisse être réduit à un apprentissage individuel par association dans la majorité des cas. Professeur Reader souligne d’ailleurs à cet effet les résultats de l’étude de Lefevre & Giraldeau sur l’apprentissage social de différentes espèces d’oiseaux, soit ceux d’une corrélation positive entre différentes tâches d’apprentissage social et individuel (l’apprentissage social ne serait donc pas une adaptation isolée, spécialisée). Dans le même ordre d’idée, la variation de l’apprentissage social est habituellement corrélative à d’autres traits tels que le comportement innovateur, l’utilisation d’outils, les tromperies tactiques…etc. soutient le professeur Reader. Ainsi, bien que l’humain par exemple, comparativement aux primates, présente une meilleure aptitude à la réalisation de tâche dans le domaine social (alors que le domaine physique est plutôt similaire), l’adaptation spécifique sur laquelle s’appuierait ce que nous pourrions nommer une intelligence sociale/culturelle accrue, m’échappe ; en quoi cela ne pourrait-il pas être le résultat de l’évolution d’un apprentissage individuel par association (l’autre = potentiellement instructeur d’un savoir -- le savoir = profitable… )
Par ailleurs, il semble que la représentation, la reconnaissance de l’autre tel qu’un autre individu et non pas un stimulus parmi tant d’autres puisse faire l’objet d’une adaptation spécialisée. En ce sens, l’intelligence culturelle transcenderait le mécanisme d’apprentissage individuel/associatif ; percevoir l’autre tel qu’un autre permettrait possiblement une meilleure coopération, entraide, troc de savoir et donc augmentation du fitness collectif que si l’autre n’était pour moi qu’un signal potentiel disponible dans mon environnement.
La position parcimonieuse du professeur Reader lui permet d’aborder et de comprendre l’apprentissage social dans son apparente complexité chez l’humain, comme le résultat de l’évolution d’un processus ordinaire aux conséquences extraordinaire. Est-ce cependant « seulement » le langage qui permet une si grande effectivité de ce type d’apprentissage chez l’espèce humaine? Quelles sont les variables responsables d’une telle évolution?
Mirror neurons are thought to be involved in social cognition, both in learning processes and in emotional processes, including empathy. Hypotheses have been put forward and suggest that neuronal dysfunction may be involved in the difficulties of putting people into contact with others in autistic people. Malfunctions of mirror neurons could be the cause or consequences of autism. Would you know if recent research supports these assumptions?
ReplyDeleteThank you.
Is there a general rule for how long it takes for observers to pick up and learn a new strategy from a demonstrator?
ReplyDeleteThank you for the presentation. In the segment about social learning of alarm responses, you mentioned that one of your students investigated the differences in brain response between the groups exposed to the aversive odor and the ones who were exposed to the non aversive odor. What were the differences in brain activity? And why would there be if the behaviors are similar?
ReplyDeleteAs Pr Reader and his colleagues seem to be polyvalent researchers working with or studying many very different species from primates to fishes, I am curious to know why there is not very much interest or studies about the raccoon species among the researchers interested in innovation in general. It seems to me they are some kind of champions in innovation and often social learning too? In Toronto, many raccoons learned very rapidly to open carefully "raccoon-proof" designed garbage bins (https://toronto.ctvnews.ca/video?clipId=1369196), another one found how to manage hiding in a man hole (http://www.asahi.com/ajw/articles/AJ201710170054.html, how to escape and avoid a trap (https://www.youtube.com/watch?v=IuRgiG7HY5Q) and those who like me have experience with them can see they often bring "friends" (ie other adults)when finding a new source of food (https://www.youtube.com/watch?v=OBXbfpP2vDM).
ReplyDeletehttps://link.springer.com/article/10.1007/s10071-017-1129-z
http://www.apa.org/monitor/2010/11/raccoon.aspx
Si l'apprentissage par le social est aussi répendu, pourquoi ne voit-on pas plus d'évolution au sein des espèces? Qu'est-ce qui limite l'innovation des autres espèces? J'imagine que leur corps est une limite importante de leur possibilité, mais il me semble que la capacité d'apprendre par le social devrait mené à plus d'évolution rapidement? Les espèces capables d'apprendre par le social devraient être en mesure de développer des capacités plus complexes. Est-on témoin de limites évolutives? Chez des espèces sous-marines qui sont présentes depuis très longtemps par exemple? Si c'est le cas, peut-on dire qu'elles sont parfaitement adaptées à leur environnement? Cela me semble impossible.
ReplyDelete