Friday 21st June: Claudio Tennie (Tübingen) Action copying in humans and other apes (and the last common ancestor) (14.00-15.30 in S1.50)
An ability to copy actions is fundamental for evolving pure action-based cultural behaviour, such as dances and rituals. While recent findings have shown that (at least some) technology can also be transmitted without action copying, action copying has arguably played a large role in technological evolution. Action copying is therefore a key mechanisms of cultural evolution, and any species capable of copying actions should therefore show the signature of cultural evolution: the cultures of such species should quickly diversify across time and space – to the point that naive subjects of the species should be unable to reinnovate culturally evolved items. To illustrate this, consider that no naive human could reinnovate an Iphone or the dance steps of tango on the spot (history, as a natural experiment, proves this to be the case). There is not much archaeological data for non-human ape (henceforth apes) culture to study the time-depth aspect of this signature, but the little data that exist suggests that ape culture did not change substantially in thousands of years. There is much more data on ape culture across space – i.e. across populations. This data however paints the same picture: generally speaking, ape behaviour repeats itself across unconnected populations – suggesting that apes can reinnovate the underlying behaviours on the spot. Indeed, when we provided naive apes with the necessary raw material, but without providing them demonstrations of the target behaviours, these apes readily reinnovate these targets, i.e. they spontaneously showed the wild form of the behaviour individually. In sum, ape culture does not show the signature of cultural evolution. This allows for only one conclusion: contra popular belief, wild ape behaviour (their culture) does not point towards action copying skills in apes. In additional tests (where only action copying can lead to success) we established that apes indeed fail to spontaneously copy actions. The general conclusion we can draw from this data is that not only apes, but also the last common ancestor of humans and apes did not copy actions. This conclusion also fits with the finding that the earliest evidence of technology in our lineage (stone tools) lacked a cultural evolution signature for millions of years. Action copying arrived late in our lineage.
Tuesday 11th June: Bert Timmermans (Aberdeen) Social interaction: from joint attention and motor learning to gaze dynamics and social agency (16.00-17.00 in The Cowling Room: S 2.77)
How do we make sense of others? Social cognition is most often studied in isolation, whereby people are exposed to social stimuli that they perceive, and which influence their actions and thoughts. This has given rise to the notions of Theory of Mind and simulation theory, but, while we undoubtedly have the capacity to reason about others and automatically take others’ perspective, most of our social world consists of interaction, which brings something unique to the table: rather than merely seeing other’s actions, I can see the effect of my actions upon other people. In this talk I will discuss a number of experiments that look at how experiencing the contingencies between my own and others’ reactions shapes how we experience others, and also ourselves. Specifically, I will focus on four questions: (1) How does initiating of joint attention shape how I see others? (2) Do such action contingencies, in eye gaze and gesture, carry an intrinsic reward that can drive my motor learning? (3) Beyond action contingencies, does a dyad of people exhibit dynamics that can predict joint behaviour? (4) If another agent’s reaction is the effect of my action, does that influence my experience of my own agency?
Thursday 6th June: Richard Moore (Berlin School of Mind and Brain) Pragmatics Made Simple (14.00 in H2.03)
An ongoing debate in language evolution research concerns the extent to which pragmatic phenomena – agents acting with and attributing communicative intentions – play a role in the evolution of language. Some (Tomasello, Sperber and Wilson, Scott-Phillips) hold that language could not be acquired independently of the ability to attribute communicative intent. Since their view makes attributing communicative intentions cognitively demanding, it entails that language-like phenomena must be phylogenetically recent and uniquely human. Problematically, such views seem inconsistent with empirical data about both the communicative and socio-cognitive abilities of non-human great apes and young children. Others (Bar-On, Planer, Sterelny) argue that understanding communicative intentions may not be necessary for acquiring language. However, without pragmatic understanding, it is unclear how key steps in language development can be explained.
I sketch a minimal view of pragmatic phenomena that may be both phylogenetically old and adequate to explain human language development in ontogeny. In the first half of the talk, I summarise my work on ‘minimally Gricean’ communication. In the second, I develop the idea that the contents of utterances can be inferred using a simple prediction-error feedback system. I finish by considering the nature of the pre-verbal cognition that we must attribute to communicative agents to explain their language development, particularly with respect to Apperly and Butterfill’s claims about System 1 mindreading.
6-10 May: Visit from Jakob Hohwy (Monash University)
Thursday 2nd May: Thom Scott-Philipps (CEU, Budapest) Perspective taking is spontaneous but not automatic (16.00-17.00 in H1.48)
Data from a range of different experimental paradigms — in particular the dot perspective task — have been interpreted as evidence that humans automatically track the perspective of other individuals. Results from other studies, however, have cast doubt on this interpretation, and some researchers have suggested that phenomena that seem like perspective-taking might instead be the products of simpler behavioural rules. The issue remains unsettled in significant part because these different schools of thought, with different theoretical perspectives, implement the experimental tasks in subtly different ways, making direct comparisons difficult. Here, we explore the possibility that subtle differences in experimental method explain otherwise irreconcilable findings in the literature on the dot perspective task. Across four pre-registered experiments we show that the classic effect in these studies is not automatic (it is not purely stimulus-driven), but nor is it exclusively the product of simple behavioural rules that do not involve mentalizing. Instead, participants do compute the perspectives of other individuals rapidly, unconsciously and involuntarily, but only when attentional systems prompt them to do so (just as, for instance, the visual system puts external objects into focus only as and when required). This finding prompts us to clearly distinguish spontaneity from automaticity. Spontaneous perspective taking may be a computationally efficient means of navigating the social world.
Monday 29th April: Kami Koldewyn (Bangor University) Perceiving Social Interactions in the Social Brain(12.00-13.00 in H4.03)
Humans are inherently social and our understanding of the world is shaped from the very beginning by the social interactions we observe and engage in. As a consequence, we excel at extracting information from social scenes. Social interactions are multifaceted and subtle, yet we can almost instantaneously discern if two people are cooperating or competing, or flirting or fighting. We swiftly learn a great deal about people from observing their interactions with others – even a brief interaction gives us important clues about their personality, their social abilities and their current mood. Here, we explore the brain basis of this remarkable ability, in particular looking at the role of various structures in the “social brain”. In a series of experiments, using fMRI, we show that a region in the superior temporal sulcus, identifiable in the majority of subjects individually with a short functional localizer scan, responds about twice as strongly when viewing pairs of people interacting with each other compared to pairs of people acting independently. This heightened response to seeing social interactions is unlikely to be accounted for in terms of simple perceptual features because the same region responds more to interactions than independent actions whether the agents are people depicted in video clips, people in point-light displays, or simple animated shapes. This region also shows sensitivity to the content and/or valence of the observed social scene. The response and development of this region will be explored in particular, but will be also be discussed in the context of response in other parts of the social brain.
Thursday 25th April: Sophie Milward (Portsmouth University) Transmission of and interference from others’ cognitive states: evidence from human children and great apes (16.00-17.00 in H148)
Human adults automatically co-represent the task of a co-actor online when acting in a minimally joint manner (Sebanz et al., 2003). This mechanism is argued to have evolved to aid prediction of a joint action partner, but it also causes interference on one’s own task performance in certain circumstances. I will present a series of studies looking at the ontogeny and phylogeny of this phenomenon. Study 1 looked at how co-representation develops over childhood, with results suggesting that this is a fairly late-developing phenomenon (around 4 years). Study 2 looked at why co-representation might develop at this age, by measuring individual differences in children’s co-representation interference in relation to other developing cognitive skills such as executive functions and Theory of Mind. Study 3 introduces an evolutionary perspective, investigating whether this tendency is a human-specific adaptation (possibly for collaboration) or whether it is shared with great apes. I am currently expanding this investigation of task interference to transmission of/interference from other cognitive or emotional states, and hope to present some preliminary data from a new project on stress transmission from adults to children.
Wednesday 17th April:Mini Workshop (15.00-18.00 in MB0.08)
Joulia Smortchkova (Oxford) First Impressions
The topic of my presentation is first impressions of others: these are, for example, the impressions that we (in most cases automatically) form when we enter a room full of people and quickly judge who looks friendly or approachable. Such impressions are very similar to judgements based on perceptual experiences, and they can be used in the inferential processes we use to think about others. In my presentation I will discuss their possible role in understanding others in connection with recent debates about the plurality of social cognition.
Jake Quilty-Dunn (Oxford/ Washington University, St Louis) Concepts as generative pointers
This talk argues for a novel theory of concepts. Standard approaches in cognitive science take concepts to be structured bodies of information that underlie categorization and inference (such as prototypes or theories). However, the heterogeneity of such representational structures undermines this approach. Some try to cobble together heterogeneous concepts via “hybridist” (Vicente & Martínez Manrique 2016) or “pluralist” (Weiskopf 2009) views, but this talk argues that these approaches fail. Another option is Fodorian atomism, on which the concept C is an unstructured vehicle with a denotation. One problem for atomism is its lack of interaction with the informational structures that are known empirically to underlie categorization and inference. Another problem is the phenomenon of polysemy, wherein single word meanings (and therefore single concepts) can shift their denotations. This talk argues that concepts function as “generative pointers”–i.e., unstructured atoms that point to locations in memory where bodies of information are stored, and the selection of particular subsets of this information can modulate the denotation of the token deployment of the concept. This theory interacts with recent developments in lexical semantics, and provides a form of atomism that moves away from a rigid view of conceptual and semantic reference and takes a less polemical view toward the cognitive science of concepts than classic Fodorian atomism.
Wednesday 3rd April: Cathal O’Madagain (École Normale Supérieure) Developmental Origins of Objectivity (15.00-16.00 in S2.81)
A concept that we repeatedly lean on as scientists is that of objectivity: the idea that the facts are independent of anyone’s beliefs. This is what motivates us to look for evidence that might confirm or disconfirm a hypothesis, and why we have a ‘reliability coder’ double-check our observations; it might even be said that science is unintelligible without invoking the concept of objectivity. What is the developmental origin of this concept? Here I explore the development of false belief understanding – a key component of the concept – and raise challenges for whether tests developed so far should be thought to really get at an understanding of objectivity. I then present some new studies that aim to get closer to a general test for objectivity – including a test that requires participants to understand that ‘anyone can be wrong’ about the facts. I end with some considerations for how language and social interaction may play a role in the development of this concept.
Wednesday 13th March
Katheryn Edwards & Jason Low (Victoria University of Wellington) Characteristics of Efficient Mindreading in Uncharted Settings
Butterfill and Apperly (2013) proposed that human beings can quickly track others’ belief by using a minimal model of the mental that represents registration as a belief-like state. This insight, anchored in terms of how humans’ online social-cognition is rudimentary and its restrictions are manifest in terms of signature limits, has revolutionized scientists’ thinking about the architecture and processes required for mindreading that is cognitively efficient. To test the proposal, researchers have focused on charting the automaticity of efficient mindreading and extent to which a minimal model of the mental struggles with tracking false beliefs about object identity in the numerical sense. Scholars in New Zealand are attempting to push the boundaries of Butterfill and Apperly’s theoretical and empirical work by delineating the representations underpinning efficient mindreading in less-trodden areas. In this joint presentation, Katheryn Edwards will first discuss research investigating whether and to what extent under-specification of differences-in-perspective may be the elemental signature limit operating on efficient mindreading. She will present a new visual-detection paradigm that weaves together belief-attribution and perspectivization, and discuss data showing that adults’ reaction times in a perspective-tracking context is helpfully speeded by a bystander’s irrelevant belief when tracking two homogenous objects but not in a perspective-confrontation context when tracking a single heterogeneous object. Jason Low will then discuss work exploring potential interactions between motor representations of action outcomes and efficient belief-tracking. He will present on a new behavioural paradigm revealing that adults’ eye movements are initially modulated by motor representations whereupon observation of motor cues from the shaping of an agent’s hand (precision versus whole-hand grasping) activates observers’ corresponding motor plans, but that later modulation in action anticipation reflects some interaction between motor processing and belief-tracking so that complementary eye-gaze responses are generated. Edwards and Low will discuss opportunities for theoretical growth that the data pose to Butterfill and Apperly’s minimal model of human mindreading, and which could be the most promising research avenues to pursue in the future.
Chiara Brozzo (University of Tübingen) Action understanding as minimal mindreading
Action understanding is the process by means of which an individual identifies an outcome to which an observed series of bodily movements are directed, where an outcome is a possible or actual state of affairs. How can action understanding be mindreading, i.e. mental state ascription? The best known answer to this question invokes intention ascription, based on the idea that intentions represent or otherwise specify outcomes. When action understanding involves intention ascription, it is full-blown mindreading. I will show another, hitherto unrecognised way for action understanding to be mindreading: when it involves the ascription of what I shall call proto-intentions. These are minimal versions of intentions that I shall introduce and differentiate both from standardly conceived intentions and from the mental states that are part of Minimal Theory of Mind. Proto-intentions differ from standardly conceived intentions insofar as proto-intentions are subject to normative constraints that are context-bound, unlike those applying to full-blown intentions. Proto-intentions also differ from the mental states postulated by Minimal Theory of Mind insofar as they are representations rather than relations. When it involves proto-intention ascription, action understanding is minimal, rather than full-blown, mindreading.