Research

Our research group has a broad interest in both fundamental and applied cognitive science. Rather than try and give a comprehensive list, below is a snapshot of the projects that we worked on in 2022. Please also feel free to take a look at our publications to see what kinds of things we are interested in.

Decision Making

In one decision-making project, we are looking at the relationship between how confident one feels after making a decision and the willingness to seek out more information. There are many forms that more information can take. We are interested in the distinction between information that informs one’s own evidence accumulation process (e.g., asking to look at the existing information again), and asking for advice about the decisions itself (e.g., asking for a recommendation about the eventual choice to be made). At its core, we want to understand the different ways that low confidence can manifest – when do people decide that they can’t decide alone, and seek external advice?

In another decision-making project, we want to understand what makes different models distinct. When it comes to simple perceptual decisions (e.g., is this patch of moving dots moving coherently to the left or right), most theories share a lot of assumptions. For example, almost all theories assume that information about the potential responses is collected and accumulated until one response has sufficient evidence to warrant its choice. Within this framework of decision-making models, there are lots of variants – e.g., some assume that responses inhibit one another, while others assume independence. In one project, we are looking at what are the consequences of having different assumptions.

Reasoning with others

In this project we were interested in how advice from other people is incorporated into our own decisions. People likely use relatively crude heuristics, or rules of thumb, for integrating information from other people into their own decisions. However, the kinds of strategies that people choose to use come from their ability to understand and explain their environment (including the task they have to perform, and the role of other people).

We ask people to perform a relatively simple advice-giving game, and classify their behaviour with computational models that reflect different strategies for incorporating advice from others. We want to show that the strategy use comes from explaining the various features of the task, including the task itself (e.g., the quality of the advice given) as well as the story they are told about the advice giver (e.g., what they are told that the advice giver knows). We also ask participants to explain what they are doing in free response, and use machine-learning language models to classify their responses to see if their explicit reports are consistent with their inferred strategy use.

Implicit and Explicit Knowledge

How knowledge makes the apparent jump from influencing behaviour without awareness to having an explicit and conscious explanation is a fascinating question. The best explanation for how this happens that we know of is that an unexpected event tells you that you actually know something, and then you try and understand what you know, by explaining the unexpected event. For example, an English learner may be surprised when they quickly and correctly guess that a new plural ends with an “s”, and then, in an attempt to explain how they could have known that, realise explicitly that the vast majority of the plurals they have learned thus far also end with an “s”

In this research project, we wanted to use a sequence-learning task to trigger an unexpected event. We hoped that participants would press buttons in a sequence that was so fast that it precluded an explicit awareness of the fact that there was a regular and repeating sequence. After this supposedly implicit learning occurred, we slowed the task down for a single trial, hoping that participants would then unexpectedly anticipate the next button press in the sequence. If we could ask them at that precise moment, then they should be able to tell us that they know there is a sequence, but be unable to tell us what that sequence was.

Motivation and Explanation

If you read the scientific literature, you would be left with the impression that people hate effortful thinking. And yet, most people have probably experienced the enjoyment of trying to solve some mystery or puzzle (e.g., a detective story, an escape room, a sudoku). While some of the happiness certainly comes from being correct, coming up with plausible guesses is itself often a rewarding experience, suggesting that it is the thinking itself that is pleasurable. We think the reason for the difference between typical theories and the real world is due to the nature of the kinds of thinking that is required in typical mental-effort studies and that which we enjoy. Experiments are usually repetitive or monotonous, requiring a simple strategy to be implemented many times over. Fun thinking, on the other hand, requires you to constantly make informed guesses about what is really going on, or more simply, to explain. Indeed, we think that the act of creating explanations is what people have evolved to enjoy.

To study this idea, we want to see if we can make a task more cognitively effortful, but also more interesting and fun for a participant. To do this, we make a version that needs people to conjecture or discover certain aspects of the task environment, while the alternative requires the same basic processes but everything needed is already known. People try out the different tasks, and in some experiments we then asked them which they would prefer to do for the remainder of the experiment. In other experiments, they are asked whether they would forgo money to do the more effortful task (i.e., in exchange for a more enjoyable experience).