Dr. Marta Garrido
Principal Investigator and Associate Professor University of Melbourne
Postdoctoral Fellow University College London
Postdoctoral Fellow University of California, Los Angeles
PhD in Neuroscience University College London
Dr. Marta Garrido is an associate professor and head of the Cognitive Neuroscience and Computational Psychiatry Lab at the University of Melbourne in Victoria, Australia. Her research focuses on understanding the neural mechanisms underlying learning and decision-making, with a particular emphasis on how the brain adapts to changes in the environment.
Marta first became interested in the brain as an undergraduate in engineering physics, thanks to a biophysics course taught by a particularly enthusiastic professor. She was fascinated to learn about experimental techniques capable of measuring activity in the human brain, and what can be revealed about cognition by combining these with behavior. At the recommendation of this professor, Marta started reading books on neuroscience. Her interest continued to deepen the more she learned. Looking back, Marta reflects that discovering neuroscience almost felt like falling in love. She remembers thinking: “I love this, this is what I want to do for the rest of my life”. One might have expected the honeymoon period of this romance with neuroscience to have ended by now. However, over a decade later, Marta's curiosity and passion for the brain remains just as strong.
Having decided that neuroscience research was the path for her, Marta moved to London to undertake her PhD in the lab of Dr. Karl Friston at University College London. Her PhD work focused on using computational models to understand the brain networks underpinning ‘auditory prediction error’. Consider someone attending a concert of their favourite band. If they know the music well, they will likely be able to predict how the rest of a song will sound after just a few bars. Should the song start to sound different, perhaps due to the noise of the crowd or differences in the musical arrangement, this will generate a ‘prediction error’, as the incoming sensory information no longer matches the listener’s expectation. These errors, and how they shape our expectations, are not exclusive to the auditory system. Indeed, there is evidence of this kind of predictive coding in all sensory systems. Nonetheless, understanding how prediction errors are generated in one modality may provide insight into the general principles that define how our brains create and update our internal models of the world.
As concert-goers rarely appreciate neuroscientists bringing their bulky equipment to gigs, Marta instead studied auditory prediction error in the lab using simple and well-controlled auditory paradigms in which participants were presented with a series of repeating sounds that sometimes varied unexpectedly in frequency or pitch. Marta then used techniques such as EEG and MEG — which monitor electrical and magnetic activity in the brain, respectively — and compared recorded neural signals in response to these patterns of sounds with those predicted by computational models. Ultimately, she was able to demonstrate that auditory prediction error engages a wide network of brain regions, including the superior temporal gyrus, primary auditory cortex and inferior frontal gyrus. She also showed that these prediction error responses involve feedforward and feedback interactions, indicating that perception is not just a passive registration of information from the environment, but an active process of prediction based on past experiences.
Marta describes her time in the Friston lab as an immensely positive experience. Karl was an inspiring scientist and mentor with an excellent sense of humour who taught her a unique way of thinking about the brain. Though there were times she felt intimidated being the only woman in the lab, particularly as she felt far less knowledgeable than her peers in the beginning, she quickly gained confidence thanks to the supportive and collaborative environment Karl had created. This collaborative atmosphere led to many co-authored publications, which Marta believes made her PhD enjoyable and productive. When she wrote her first paper as a PhD student, Marta recalls taking it very personally when it was initially rejected. Since then, she has realized the importance of recognising the limits of your own work. Research is never truly finished and there is always room for improvement, particularly as advances in technology and understanding can reveal gaps or misconceptions in past results. Although frustrating at times, Marta notes that “one only reaches the truth in an asymptotic way”. Even if we can’t be sure a given paper will withstand the test of time, every piece of research is still pushing us forward towards a better understanding of the brain.
After her PhD, Marta did a short postdoc at UCLA before returning to UCL to work as a postdoc in the lab of Dr. Ray Dolan. During this time, Marta became fascinated by blindsight: a phenomenon in which people with visual impairment have the ability to respond to visual stimuli without consciously perceiving them. Using a combination of anatomical data from the Human Connectome Project and functional data using fMRI, she was able to demonstrate the existence of a ‘shortcut’ in the brain between the pulvinar nucleus and the amygdala that bypasses the visual system entirely. These brain regions are known to be critical for functions such as visual processing, in the case of the pulvinar nucleus, and memory, decision making, and emotional responses (particularly fear, anxiety, and aggression) in the case of the amygdala. Marta found that people with greater fiber density in this pulvinar to amygdala pathway were better at recognising fear. Although tested in healthy participants, the existence of this pathway provides a potential route by which people with blindsight could process visual information without involving the visual system, perhaps accounting for their lack of conscious visual perception. She hopes to one day test this theory in people with blindsight, although this condition is quite rare so such an opportunity has not yet arisen.
Marta is now principal investigator of her own lab at the University of Melbourne. Continuing to build on topics she became interested in during her PhD, the main focus of her lab is to understand how humans integrate information from the environment with prior beliefs to make decisions, particularly in cases of psychiatric disease. For example, recent work in the lab has used a behavioural task to explore how people with autism and schizophrenia rely on different sources of information to make decisions in uncertain situations. In this task, participants must guess where a coin fell in a virtual pond based on the splashes and ripples it produces. Beforehand, participants are told that the coin thrower is aiming at the middle of the pond. In this task, the ripples act as a source of sensory or ‘likelihood’ information, while the history of previous throws and the knowledge that the thrower was aiming at the middle act as priors or ‘expectation’ information.
When tested in healthy controls, Marta has found that people integrate both sensory and prior information to guide their guesses but tend to rely more heavily on the source of information that most reliably predicts the location of the coin. However, when tested in people with autism or psychotic traits, she observed something different. In the case of autism, individuals appeared to have a more precise encoding of sensory information and “sticky” priors – meaning they more accurately encoded the location of the splashes but didn’t update their expectations for future throws accordingly. In contrast, in people with psychotic-like experiences, it appeared the influence of priors was weaker overall. Although more follow-up work is required, this provides an exciting starting point for teasing apart the mechanisms by which the brain makes decisions under uncertainty and what deviations in this system can teach us about psychiatric disease.
Marta’s lab has now been up and running for several years and Marta continues to take great joy in watching her lab grow and supporting her students to develop as scientists. She admits that learning to work with and manage people is both one of the biggest privileges and challenges of running a lab, and she wishes that there was more formal training and support for junior faculty starting out with their own lab. With a career profoundly shaped by the infectious enthusiasm and dedication of her mentors and professors, Marta is striving to inspire the next generation scientists through her own passion and supportive mentorship.
Find out more about Marta and her lab’s research here.
Listen to Caitlin’s full interview with Marta on December 1st, 2022 below!
