| Visual working memory(VWM)is a central cognitive function,which usually maintains multiple items for a short period of time,and then retrieves features from the remembered items.Studies have indicated that freely choosing the retrieval sequence leads to better performance of VWM compared with forced report.However,the cognitive,spatial and temporal mechanisms of the difference between free and forced conditions have not yet been systematically investigated.Thus our series of studies intended to investigate the difference between free and forced recall as well as the underlying cognitive and neural mechanisms through three experiments including behavior,magnetic resonance imaging(MRI)and electroencephalogram(EEG).Experiment 1 was a behavioral experiment,using high-temporal-resolution mouse trajectory recording as well as computational modeling to reveal the cognitive mechanism of the free recall benefit.Sixteen participants completed a VWM recall task.Three items of colored squares were presented for 200 ms.After a 900 ms delay period,participants were asked to retrieve all the three items sequentially with orders either defined by themselves(free)or randomly assigned by the computer(forced).We found participants tended to recall the item with the highest precision in both free and forced recall.During the forced recall,when the first-probed item was not with the highest precision,the highest-precision item could interfere with the retrieval and cause a deflection of mouse trajectory towards the highest-precision item.Such higherprecision-first strategy led to the better precision and shorter item-chosen reaction time in free recall.We also found that participants tended to report the top-left item first,which was not contributing to free recall advantage.Experiment 2 further explored the neural mechanism of free recall advantage using functional MRI.The cue onset in the free recall significantly activated the left pre-and post-central cortex,the Temporo-parietal Junction(TPJ)and Supplementary Motor Area(SMA),compared with forced recall.These results indicated that a prior motion plan might have been formed earlier in free recall.The cue in forced recall activated the left Angular Gyrus which might be related to inhibition of default mode network.Those regions activated in free recall had stronger functional connectivity in forced recall,suggesting a pre-planed motion was also formed in forced recall.These brain imaging results were related to behavioral performance,which was reflected by the correlation between activities in pre-and post-central as well as Angular Gyrus and RT.Experiment 3 further explored the temporal dynamics of the processes mentioned above using EEG and added confidence rating in the task.We found earlier-reported items came with higher confidence.However,when we separated metacognition into three parameters using a metacognitive model,we found no differences between free and forced recall on the three parameters of sensitivity,bias and concentration.Through a decoding analysis using EEG data,we found in free recall,spatial location could be decoded at early encoding stage while in forced recall spatial location could not be decoded.Thus the retrieval sequence might have been generated during encoding.In free recall,participants retrieved this sequence again.But in forced recall,the sequence was randomly chosen thus could not be decoded.Base on all the results from three experiments,free will advantage in VWM could have started from the encoding stage.Participants might sort the items according to their precision and generate a recall sequence during encoding of multiple items.In free recall,they usually retrieved high-precision items first while in forced recall when the probed item was not the high-precision one,high-precision items would attract mouse trajectory and prolong RT.Free recall advantage was also related to activation in the sensorimotor cortex during cue onset.In sum,free recall can free VWM and improve VWM performance compared with forced recall. |
PDF Full Text Request