Angélina Lacroix: APNM Undergraduate Research Excellence Award Winner

Canadian Science Publishing sponsors the Applied Physiology, Nutrition, and Metabolism Undergraduate Research Excellence Awards, which are awarded in partnership with the Canadian Society for Exercise Physiology and the Canadian Nutrition Society. Award winner Angélina Lacroix shares her research on how the brain responds to reward versus punishment.


By Angélina Lacroix


Last year I had the opportunity to complete two research internships under the supervision of Pierre-Michel Bernier, an associate professor from the Kinanthropology Department at Sherbrooke University in the laboratory of learning and motor control. During this time I was in charge of the experimental design and data collection for a study called Added value of monetary feedback on motor performance feedback: respective increases in motor beta-band and mid-frontal theta-band power for rewards and punishments. In addition to gaining a lot of practical experience, this internship has confirmed my passion for neuroscience research.   

Rewards and punishments: how do they affect the brain?

Throughout our lives we develop according to the experiences we face, some of which are positive and others are negative. In both cases we learn and improve our skills for facing future challenges. This learning process is the result of changes taking place in our brain that take into consideration the valence (i.e., the positive or negative nature) of the feedback received in response to a novel experience (e.g., being rewarded versus being punished). This differentiation allows us to select or develop a better strategy for similar situations in the future. To ensure that it happens, the brain must respond differently to a positive feedback versus a negative one. But what is happening in the brain during this differentiation? 

Measuring how brains respond to positive and negative feedback

It is well known that giving a reward following an action tends to reinforce the action that led to the reward, making it more likely to be reproduced in the future. In contrast, when a punishment is given it leads to the avoidance of the punished action. For example, you would probably repeat the behaviour that resulted in an excellent grade on a test (e.g., studying) and avoid what made you fail (e.g., partying). While these types of learning subtended by positive and negative feedbacks are well studied, the precise neuronal processes responsible for these behavioural changes are not fully understood. Dr. Joseph Galea and colleagues1 found that negative feedback improves short-term motor performance and, conversely, that positive feedback enhances long-term retention of motor behaviours. These results suggest that positive and negative feedbacks are mediated by distinct brain processes.

Similarly, studies using gambling tasks or time-estimation tasks (where participants guess the time a stimulus appeared or its duration) and electroencephalography (EEG), which records the electrical activity or signals of the brain, have shown that 200 to 400 ms post-feedback positive and negative feedbacks are respectively associated with specific signals called beta and theta power2-3. However, we do not know if these neural processes associated with different types of feedbacks are also at play during sensorimotor tasks (i.e., tasks that involve motor function and sensory perception).

The aim of this study was to investigate the brain activity associated with positive (reward/monetary gain) and negative (punishment/monetary loss) feedbacks during a sensorimotor reaching task. Using EEG, 23 participants were seated in front of a mirror and held a computer mouse. The mirror reflected the target from a computer monitor and participants could not see their hand or the mouse cursor (Fig. 1). We asked them to move toward a visual target with their right hand.

 
 Fig. 1: Experimental setup.

The target was one among eight possible conditions (Fig. 2). Each of the eight conditions was designed to control for key parameters, namely reaching probability and the valence (Fig. 2). To control reaching probability, the target’s sizes were adjusted according to the accuracy of each participant during a baseline assessment. Regarding valence, the colour of different zones in each target indicated the amount of money the participant would get if he/she succeeded in moving the mouse cursor to the zone. At the end of the movement the final cursor position was revealed to the participant, and based on the cursor position the monetary gain (reward) or loss (punishment) was indicated. 

 
Fig. 2: Valence conditions. 

Rewards and punishments affect different parts of the brain 

The post-movement analysis demonstrated that 200 to 400 ms post-feedback participants missing the target had a higher mid-frontal theta power. This increase in mid-frontal theta was greater when the failure was not expected (i.e., in conditions with high success probability and a larger target zone) as well as when failure was coupled with a punishment (monetary loss). Conversely, when participants successfully reached the target, beta power occurring in sensorimotor areas in the same time window was increased. This response in beta power was amplified when participants did not expect to succeed (i.e., in conditions with low success probability and a smaller target zone) and when a reward was given (monetary gain). 

The results of this study provide evidence that our brains have distinct neural processes for positive and negative feedbacks during sensorimotor tasks. The post-feedback mid-frontal theta power seems to be associated with negative outcomes or task failure. The post-feedback sensorimotor beta power seems to be associated with positive outcomes or task success. Considering these results, it is possible that modulations of the brain’s electrical activity play a major role in motor learning. These modulations are perhaps a way for distant brain areas to interact while processing both correct and incorrect feedback3.

By identifying the neural basis of rewards and punishments processing, this study may contribute to the improvement of therapeutic interventions for populations suffering from neuromotor diseases. Technologies that can modulate electrical activity, such as transcranial magnetic stimulation, could influence how we perceive rewards and punishments. In this way, increasing or decreasing theta and beta modulations could enhance an individual’s capacity to acquire and retain new movements.


   
References
1The dissociable effects of punishment and reward on motor learning
2Theta and high-beta networks for feedback processing: a simultaneous EEG–fMRI study in healthy male subjects
3High-learners present larger mid-frontal theta power and connectivity in response to incorrect performance feedback


Angélina Lacroix is a neurosciences MSc student at Sherbrooke University, Sherbrooke, QC. She conducted research with Jean-François Lepage (Faculty of Medicine and Health Sciences) and Pierre-Michel Bernier (Kinanthropology Department) during her undergrad at Sherbrooke University.