Key Findings:
- Visual experiences are constructed from comparing actual input from the eyes (input) to hypotheses about what we might see (expectations).
- Different layers of the visual cortex are responsible for processing expectations and input.
- Machine learning can be used to decode the content of a person’s visual perception.
The question of how brains produce our rich perceptual experience has fascinated philosophers and scientists alike for centuries. Dr. Peter Kok, a cognitive neuroscientist at University College London, uses cutting edge technologies, optical illusions, and the element of surprise to tackle this age-old question in the field of visual perception. For his influential empirical and theoretical contributions, FABBS is delighted to recognize Dr. Kok as an Early Career Impact Award winner from the Cognitive Neuroscience Society.
While conscious perception feels like a straightforward reflection of the environment, Dr. Kok’s research shows that visual experiences are actually constructed from both our expectations and the light that enters our eyes. Predictive coding theory, which motivates much of his research, suggests that brains infer reality through a process of comparing input from our eyes to expectations of what we might see. Importantly, these expectations can bias how we ultimately interpret reality, even leading to errors in perception, visual illusions, or hallucinations.
Cognitive neuroscience has made leaps forward in the past decade, aided by new technologies that allow for more fine-grained assessment of activity in the brain. For example, advances in the resolution of functional Magnetic Resonance Imaging (fMRI) allow researchers to observe a fraction of a millimeter of the brain at a time. This level of detail is critical for answering the types of questions that fascinate Dr. Kok, since he can localize specific cognitive processes to different layers of the visual cortex. Discerning where expectations and input are processed gives rise to a better understanding of how the brain constructs a cohesive perceptual experience, including what may be occurring when these processes go awry.
While Dr. Kok identifies as a basic researcher seeking new discoveries without immediate applications, remarkable breakthroughs in his research may ultimately reveal the mechanisms underlying clinical conditions. For that reason, he also collaborates with researchers who study diseases that cause psychosis or hallucinations. By viewing different layers of the visual cortex, they aim to discern whether a hallucination arises from overly strong or faulty expectations or alterations in the processing of input from the eyes. Funding for basic science research, including the development of new technologies and methods of analysis, can result in unanticipated but highly fruitful insights in the future.
Indeed, many of Dr. Kok’s contributions have involved innovative ways of using existing technologies to answer questions that were previously beyond the scope of science. For example, magnetoencephalography (MEG) and fMRI have traditionally allowed researchers to measure where, when, and how much activity is occurring in the brain. However, Dr. Kok also applies machine learning algorithms to these data to measure what a person is actually expecting or perceiving in a given moment. Equipped with these tools, Dr. Kok and his colleagues design experiments to manipulate key variables, such as priming visual expectations with auditory cues or showing ambiguous or completely random images. This work has shown that expectations can bias perception when images are ambiguous, such that participants may report seeing clouds of dots move in the direction they expected them to.
Dr. Kok’s training has included computer science, psychology, and philosophy, which are the key ingredients for a neuroscientist seeking to explore conscious perception. In particular, he reflects on the value of postdoctoral fellowships for expanding his research horizons with mentors working in different countries and continents. Dr. Kok also pays forward mentorship he has received by working with scholars from low resource backgrounds and underrepresented groups through the In2Research program. Exposing these undergraduates to firsthand experiences in research labs can open doors to both future careers and scientific breakthroughs. Ultimately, funding that supports the development of researchers across their education is a priority for Dr. Kok.
Over the next few years, Dr. Kok seeks to advance our understanding of how memory intersects with visual processing. To do so, he will investigate connections between the visual cortex and the hippocampus, a key region for encoding and retrieving memories. Additionally, he anticipates rapid advances in neuroimaging techniques that will open new possibilities in the field of human neuroimaging. Rather than merely observing brain activity, he could use non-invasive and temporary ultrasound stimulation to disrupt activity and observe the causal effect of different regions on perception. Other technologies under development would allow participants to move around the world or to interact with a partner during recordings. Ultimately, the future that Dr. Kok envisions is bright.
Future Directions:
- Explore the role of memory and the hippocampus in visual perception
- Partner with clinical researchers to investigate visual alterations in different disorders
- Capitalize on new technologies and advance analysis methods
