کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
6031266 | 1188730 | 2012 | 13 صفحه PDF | دانلود رایگان |

Recent studies have demonstrated that retinotopic cortex maintains information about visual stimuli during retention intervals. However, the process by which transient stimulus-evoked sensory responses are transformed into enduring memory representations is unknown. Here, using fMRI and short-term visual memory tasks optimized for univariate and multivariate analysis approaches, we report differential involvement of human retinotopic areas during memory encoding of the low-level visual feature orientation. All visual areas show weaker responses when memory encoding processes are interrupted, possibly due to effects in orientation-sensitive primary visual cortex (V1) propagating across extrastriate areas. Furthermore, intermediate areas in both dorsal (V3a/b) and ventral (LO1/2) streams are significantly more active during memory encoding compared with non-memory (active and passive) processing of the same stimulus material. These effects in intermediate visual cortex are also observed during memory encoding of a different stimulus feature (spatial frequency), suggesting that these areas are involved in encoding processes on a higher level of representation. Using pattern-classification techniques to probe the representational content in visual cortex during delay periods, we further demonstrate that simply initiating memory encoding is not sufficient to produce long-lasting memory traces. Rather, active maintenance appears to underlie the observed memory-specific patterns of information in retinotopic cortex.
⺠Memory encoding of visual low-level features recruits retinotopic areas. ⺠V1 shows lower levels of activity when memory encoding is interrupted. ⺠Higher-tier areas encode information independent of featural content. ⺠Active maintenance is necessary for successful decoding of memory content. ⺠LO2 does not represent orientation information during maintenance.
Journal: NeuroImage - Volume 63, Issue 1, 15 October 2012, Pages 166-178