The University of Queensland

Perceptual Metacognition
People experience levels of confidence when they make decisions - and this includes perceptual decisions. Importantly, people tend to feel more confident when they have made a good decision. This degree of insight people have, into when they have made a good or a bad decision, is called metacognitive sensitivity.

Metacognitive sensitivity may be important in our daily lives, helping us to learn from our mistakes. So, it would be good to have a reliable measure of it. However, we have shown that currently popular metrics of metacognitive sensitivity suffer from interpretive ambiguities, as differences can equally be caused by true differences in metacognitive sensitivity, or by the researcher having inadvertently committed to an assumption that is wrong for the given context.

We have now developed a more robust way to measure metacognitive sensitivity, and you can find a description of that process and download code from our website.


Aphantasia and Mental Imagery
Most people can conjure images and sounds that they experience in their minds – such that we can refer to the mind’s eye and to the mind’s ear. This capacity is so general that you might be surprised that some people have no such capacity. People who are unable to voluntarily form mental images are known as aphantasics.

Two of our lab members (Derek Arnold and Loren Bouyer) are aphantasics. Derek cannot have imagined visual experiences, and Loren cannot have imagined audio or visual experiences. Indeed, she does not even have visual or auditory experiences when she dreams. We are therefore very interested in working out what features of brain activity allow some people to have vivid imagined sensory experiences, while other people cannot.

We have a number of goals in this ongoing project.

1) We want to develop and validate metrics of the vividness of imagined sensory experiences that are more accurate than self-report questionnaires. Some candidates measures exist, and we will be testing these and striving to develop our own measures.

2) We want to develop and validate EEG measures of brain activity that predict the subjective intensity of imagined sensory experiences. At the moment, there are some fMRI measures that predict the subjective intensity of imagined experiences, but there are no validated EEG measures.

3) Long term, we would like to determine if aphantasics represent one end of a spectrum that encompasses the general population, or if aphantasics are qualitatively distinct from the general population.


Predictive Processes in Human Vision
Prediction is a key function of the human brain - and of the human visual system. Without predictive processes, you would never be able to catch a flying ball, or dodge moving objects.

Our brain activity is delayed relative to the real-world events to which it relates. So, if the brain activity that represents the positions of moving objects were simply updated to reflect contemporary input, we would always see moving objects in positions that lag their true positions in the external world.

Clearly the processing delays of the human visual brain are compensated for, by the processes that combine to promote our ability to intercept and avoid moving objects. But is this achieved exclusively by motor planning, or do our brains implement predictive computations that make us see moving objects in their true contemporary positions?

The processes underlying visual predictions are an ongoing point of interest. In recent studies...

1) We have shown that visual predictions are more accurate when input conforms with our internal model of the world.

Our internal model of the world encompasses an understanding of how gravity should impact on the movement of flying objects. We linked this and successful predictions to enhanced alpha-band oscillatory visual brain activity. This may be a marker of a greater capacity to attend to task relevant information when our predictive models of the world are accurate.

2) We have shown that when people make explicit predictions about what visual input they will next encounter, their visual brain is more responsive to a predicted than to an unpredicted input.

This finding is in contrast to most studies that consider how prediction shapes the brains' responses to inputs. Many studies have encouraged the formation of implicit expectations, by manipulating the probability of encountering different inputs. When this is done, human brains tend to be less responsive to likely inputs, which are presumed by researchers to have been predicted by the brain.

Our finding shows that explicit predictions are different. Instead of being less responsive to a predicted input, people's brains are more responsive to an explicitly predicted input. We believe this is due to people seeking out information that confirms an explicit expectation.

3) In a related study, we found that the perceived duration of brief visual events depends on whether the event has been explicitly or implicitly predicted.

Explicitly predicted events seemed to have a relatively protracted duration, whereas highly likely events that may have been implicitly predicted seemed to have a shorter duration.