Ataseven, N. | bioRxiv Science

bioRxiv2026

Mentally transformed representations in memory are linked to their originals

Working memory is an active workspace for manipulating encoded information. Neural decoding studies show that following mental transformation, the encoded original remains represented alongside the transformation product despite offering little to no functional merit. To test why, we employed multivariate EEG-decoding while participants mentally rotated a memorized orientation grating, encoded an additional grating, and were finally retro-cued whether the rotation product or the additional item would be probed. This tested whether retention of the original reflected (1) perceptual encoding, (2) deep encoding and persistent storage, or (3) a link to the transformation product. The original was decodable after rotation, and encoding an additional item, suggesting its retention was not merely perceptual. Next, contingent on the retrocue, the original and rotation product were retained or dropped together: the original remained decodable only when the rotation product was selected. Therefore, the original did not persist because it was deeply encoded; its retention depended on attentional selection of the transformation product. Yet, we found no evidence that retaining the original was beneficial. Our results show that mental transformation links the transformation product to its source representation, reflecting a dependent representational structure. We speculate that this structure may afford behavioral flexibility in dynamic environments.

neuroscience↗

bioRxiv2026

Diffuse predictions stabilize and reshape the neural code during working memory encoding

Predictions can alter working memory (WM) representations. However, its effects may have been mischaracterized due to the use of precise predictions in previous experiments, where exact properties of upcoming memory items are cued in advance. Here we investigated a more ecologically valid scenario, in which we assessed the impact of diffuse predictions, where advance cues provided only partial knowledge about the targets. To investigate the resultant nature of the target representations in WM, we performed a series of multivariate analyses of EEG data. Forty participants judged whether a probe grating was rotated clockwise or counterclockwise relative to a memorized orientation, which was either predictable or unpredictable. Each memory item was preceded by a central color cue (red, green, or blue). In half of the trials, two of these (predictive) colors cued two non-overlapping 90{degrees} segments of orientations that the grating was sampled from. Thus, participants knew the range of possible orientations of these items, but not their exact orientation. In the other half of the trials, a third (non-predictive) color was presented, signaling that the item could have any possible orientation. Behavioral results revealed higher accuracy for predictable items, with systematic biases toward the center of the cued segment. EEG results revealed equally successful decoding of orientation for both predictable and unpredictable items during memory encoding. However, cross-condition decoding was significantly weaker than within-condition decoding, suggesting that the encoding format changed between conditions. Representational similarity analysis showed higher similarity between predictable items, with a representational bias towards the cued segment. Covariance matrices showed lower variance for predictable items while the representational space of predictable items was shrunk. These effects were absent during the maintenance phase. Together, our findings suggest that diffuse predictions alter the geometric layout of the neural representations and stabilize the neural code during WM encoding.

neuroscience↗