Current Events

There are two types of systems: complex systems and what I shall call logic sytems. Logic systems are those which have properties which enable predictive mathematical models. Most systems in reality, however, are complex. Complex systems have properties which prevent causal mathematical modelling of the entire system. Complex systems are able to adapt to changing environments and are very stable, with intermittent bursts of change. The human mind-body-continuum is a complex system. All experimental sciences dealing with complex system such as humans have a massive replication problem. The problem is that many hypotheses concerning strong effects – the low hanging fruit, in other words – those which can be confirmed with a simple experimental setup and a limited number of subjects, have already been formulated and proven. Our ability to prove hypotheses dealing with small effects, in contrast, is severely limited by the difficulties in obtaining adequate measurements, the high variance of human populations and the multi-factorial, complex, overlapping and interacting causation factors we are facing. The main underlying reason for this phenomenon is that complex systems often produce effects which yield non-ergodic distributions from which we cannot derive adequate samples. It is time to bite the bullet: Cognitive science should in the future focus on description and interpretation, but not on causal modelling or exact prediction. Policy makers using results from cognitive science need to understand how to value and utilise them.

Language production is a difficult process. Producers have at their disposal a vast (potentially infinite) array of possible names and descriptions when they set out to convey a message when writing, signing, or speaking, alone or in dialogue. In the case of choosing between closely related words that differ primarily in their length, such as chimp and chimpanzee, a number of factors are at play. Under some accounts of language production, namely information theoretic approaches, producers tailor the selection of their words depending on how surprising what they are about to say is, with prior corpus and experimental work supporting these claims. However, in a recent set of studies I have tested another account in which lexical choice does not explicitly optimize the wordforms that people select, but rather is constrained by more global factors, such as the formality of the register in which a word appears. Using cutting edge computational approaches from natural language processing in conjunction with machine learning methods, I will show that we can explain both individual behavior -- judgments of appropriateness in context -- and corpus data. The results suggest a high-level role of the broader utterance on lexical choice than information theoretic approaches have historically admitted. Toward the end of the talk, I will discuss consequences for computational modeling of language production, and specifically highlight the need for modelers to think about contextual factors such as register or socioindexicality that drive utterance planning.

Much of the research on speech production has been dedicated to understanding neural control for the muscles of articulation. This project expands on questions of neural control to investigate the potential beneficial effects of sustained oromotor activity on activity-induced cortical excitability and speech motor learning. The rationale for this line of inquiry was derived from the physical therapy literature demonstrating the beneficial effects of limb motor activity on cortical excitability and rehabilitation outcomes of upper extremity function. To our knowledge, the neuroplastic changes induced by oromotor behaviors has, in contrast, not been investigated within a framework of endogenous neuromodulation where the primary premise is that specific motor activities create an optimal neurochemical environment for learning to occur. We used a variety of methodologies to investigate the impact of sustained oromotor activity on speech production (i.e., a perceptual paradigm, speech acoustics, and optical motion capture) and neurophysiology (i.e., transcranial magnetic stimulation). The findings of this work, which I will discuss in detail during this talk, have potential for future applications in the rehabilitation of patients who are experiencing speech difficulties, by priming the brain to maximize response to therapy. Understanding how activity-induced cortical excitability is affected by oral motor behaviors in the intact central nervous system is critical for supporting maintenance of speech communication and oral intake across the lifespan.

Imagine encountering a novel fruit. Even though you have never seen it before, you will be able to make a number of predictions about it: it may be edible, it may have seeds, and the tree it’s growing on probably has more fruits like it. Moreover, other trees like this one probably grow the same fruit as well. The capability for abstraction allows humans to learn and generalize quickly from sparse data, allowing us to make wide ranging predictions in new situations. Previous research has suggested that humans may be the only species capable of abstract knowledge formation, but this remains controversial, and there is also mixed evidence for when this ability emerges over human development. I will present two studies investigating abstract rule formation in capuchin monkeys, chimpanzees and children, in experiments guided by the predictions of a hierarchical Bayesian cognitive model. In a third study, we investigated children's and capuchin monkeys' ability to integrate causal and social information when copying a goal-directed behaviour. Time permitting, I will touch briefly on a final set of studies, examining young toddlers’ and domestic dogs’ understanding of fundamental causal principles, such as temporal priority—that causes must precede their effects. Together, these studies introduce the approach of hierarchical Bayesian modelling to the field of comparative psychology as an additional method for evaluating competing hypotheses.

Much of the research on speech perception has looked at relatively “easy” listening conditions:  adult listeners, listening to their native language spoken by an unaccented speaker, with relatively little background noise.  My research is focused on how processing changes when things get harder: when there is noise and other distractions, when the speaker has an accent or switches languages mid-sentence, and when the listener has less existing background knowledge (such as is the case for both young children and our canine companions). This talk represents a brief overview of recent results from my lab that cover a range of these topics, focusing primarily on listening performance in young children and dogs.

When we talk, we constantly use our hands to make meaning. Gesture has been shown to be a key resource in a variety of teaching and learning processes. However, although a number of studies have investigated teachers’ and students’ independent use of gesture, we know less about how “conversations of gesture” unfold in classrooms. How do teachers and students respond to, take up, and build on one another’s gestures in STEM educational contexts? In this talk, I present new findings on how gestures can serve as powerful dialogic resources that are re-used and transformed to co-construct STEM knowledge in whole-class discussions, one-on-one tutoring, and small group work. Data are drawn from secondary computer science, secondary mathematics, and undergraduate physics contexts. Implications for teaching and learning STEM will be discussed.