Existing function in primary visual cortex is not perturbed by new skill acquisition of a non-matched sensory task

Abstract

Acquisition of new skills has the potential to disturb existing network function. To directly assess whether previously acquired cortical function is altered during learning, mice were trained in an abstract task in which selected activity patterns were rewarded using an optical brain-computer interface device coupled to primary visual cortex (V1) neurons. Excitatory neurons were longitudinally recorded using 2-photon calcium imaging. Despite significant changes in local neural activity during task performance, tuning properties and stimulus encoding assessed outside of the trained context were not perturbed. Similarly, stimulus tuning was stable in neurons that remained responsive following a different, visual discrimination training task. However, visual discrimination training increased the rate of representational drift. Our results indicate that while some forms of perceptual learning may modify the contribution of individual neurons to stimulus encoding, new skill learning is not inherently disruptive to the quality of stimulus representation in adult V1.

Learning alters neural activity to simultaneously support memory and action

“…1f). Reversion has been observed in various different contexts, such as reaching tasks (Perich and Miller, 2017, Cherian et al, 2013), BCI tasks in visual cortex (Jeon et al, 2022), and in the remapping of hippocampal place fields (Alme et al, 2014). This would indicate that the neural activity we observed in M1 during performance of a task can be unaffected by an intervening learning experience.…”

“…It could have been that motor memories were stored in a manner that is not detectable when another action is being performed (Herzfeld et al, 2014, Jeon et al, 2022), nor be present in the same neural activity when it is driving behavior. For example, in our experiments, the motor memory could have been stored (perhaps outside of M1) such that the memory is only detectable in M1 the appropriate behavior is being performed.…”

“…Here we overcome this difficulty by leveraging a brain-computer interface (BCI) paradigm (Jarosiewicz et al, 2008, Ganguly and Carmena, 2009, Koralek et al, 2012, Hwang et al, 2013, Sadtler et al, 2014, Gulati et al, 2017, Jeon et al, 2022, Oby et al, 2019). A key advantage of a BCI for the study of motor memory is that the relationship between neural activity and behavior (termed the BCI map) is specified by the experimenter (Golub et al, 2016, Orsborn and Pesaran, 2017).…”

Cognitive Escape Reinforcement Learning for Complex Decision Making

“…Compared with RL, human intelligence has a superior cognition ability and higher flexibility. The human brain processes the sensory information from the external environment through the sensory system to obtain abstract semantic information [41][42][43] . Then, the cognition control system, namely, the prefrontal cortex (PFC), integrates semantic information and knowledge from the memory to make decisions [37][38][39][40] .…”

“…Investigations have found that human beings have a high-level cognitive control module in the brain to get rid of the local optimum [37][38][39][40] and make a better decision and that human beings can deal with various information from different scenarios since their brains abstract primary sensory information into high-level semantics for decision-making [41][42][43] , rather than dealing with numerous low-level information directly. The superior cognition ability and higher flexibility displayed by humans can enlighten to open up new avenue for addressing the above-mentioned challenging problems of RL.…”