Influence Of Training On Working Memory And Its Prefrontal Neuronal Mechanism

Working memory is the ability to maintain and manipulate information in mind,over a time span of seconds.It has been considered to be a core component of higher cognitive functions.The storage capacity of working memory is limited but can be expanded by training.However,the neuronal mechanisms underlying the improvement of working memory capacity is still unclear.After such training,performance increases not only for the trained tasks but also for working memory tasks that were not part of the training.However,the neuronal mechanisms underlying the transfer of training effects across tasks is still unclear,too.Part ?.Visuo-spatial working memory advantage for same hemifield stimuliWorking memory capacity,the amount of information that may be maintained in mind over a period of seconds,is extremely limited,to a handful of items.Some evidence exists that the number of visual items that may be maintained in memory over a period of seconds is independent for the two hemifields.To test this assumption,we trained monkeys to perform visual working memory tasks that required maintenance in memory of the locations and/or shapes of 3-5 visual stimuli.We then tested whether systematic performance differences were present for stimuli concentrated in the same hemifield,vs.distributed across hemifields.We found little evidence to support the expectation that working memory capacity is independent in the two hemifields.Instead,when an advantage of stimulus arrangement was present,it involved multiple stimuli presented in the same hemifield.This conclusion was consistent across variations of the task,performance levels,and apparent strategies adopted by individual subjects.This result suggests that factors such as grouping that favor processing of stimuli in relative proximity may counteract the benefits of independent processing in the two hemispheres.Our results reveal an important property of working memory and place constraints on models of working memory capacity.Part ?.Neuronal activity in prefrontal cortex represent visuo-spatial working memory capacityThe amount of information that can be stored in working memory is limited but may be improved with training.Strategies such as grouping and chunking allow subjects to group multiple stimuli into fewer items that can be more easily remembered.The basis of improved efficiency at the level of populations of single neurons is unknown.To investigate this question,we trained monkeys to perform a working memory task that required memory for multiple(1-5)stimuli.The performance was lower for larger numbers of stimuli and improved with training.Neural activity in the prefrontal cortex decreased as training progressed and working memory capacity improved.However,delay period activity relative to the baseline increased as training progressed.Our results reveal that improved efficiency of working memory representations is achieved through a higher signal-to-noise ratio of delay period to baseline activity.Part ?.Plasticity of prefrontal neuron activity during learning in a working memory taskTraining to improve working memory is associated with increased BOLD signal from the prefrontal cortex.The neural substrate underlying these changes is poorly understood.In a previous series of experiments,we identified increases in prefrontal activity after training monkeys in working memory tasks.However,these experiments provided only snapshots of changes at the start and end of training.To assess how brain changes are affected by different aspects of training,we recorded neuronal activity with a chronic 64-electrode array in the prefrontal cortex of 2 monkeys,as they learned to perform a visuo-spatial working memory task.A total of 4537 multi-unit activity(MUA)records were selective for the location of the stimuli.MUAs obtained at each subsequent stage of training were characterized by increased firing rate during the stimulus presentation and the delay period.Interestingly,increased firing rate and the number of selective MUAs were also observed during the fixation task.The results also demonstrate that activity changes induced by training transfer between stimuli and task conditions.Part ?.Neurons in prefrontal cortex encode effort-based decision makingThe rat medial prefrontal cortex(m PFC)implicated in effort-based decision making,while the neuronal mechanism underlying it is still unrevealed.Moreover,the choice options for animals in these experiments were extremely limited.We investigated this issue by using the “Do more,get more”(Dm-Gm)behavioral task.A rat initialized every single trial by nose-poking in a hole and decided by themselves how much effort to invest by holding a nose-poking-in status,then they earned water reward which was positively correlated with its invested effort provide the holding duration exceeded 800 ms.We recorded 1198 event-related neurons in the m PFC.Among these neurons,some fired higher when rats executed small effort,meanwhile other neurons showed more excited activity when rats perform big-effort trails.Over all,our results strongly suggested that m PFC neurons encoded effort-based decision making...