New questions on the hemispheric encoding/retrieval asymmetry (HERA) model assessed by divided visual-field tachistoscopy in normal subjects.

According to the hemispheric encoding/retrieval asymmetry (HERA) model, based on data obtained through functional neuroimaging, the left and right prefrontal cortices are preferentially, and, respectively, involved in long-term episodic memory encoding and retrieval. In this study, the HERA model was tested from a behavioral perspective using divided visual-field tachistoscopy. A recognition paradigm with both verbal and visuospatial materials was devised to differentiate memory-related effects (encoding vs. retrieval) from effects linked to the materials. The paradigm used lists of 12 and four items to assess long-term episodic memory and short-term memory, respectively. The aim of the latter condition was to test whether the HERA model is applicable in short-term memory. For long-term episodic memory, the data obtained validated the HERA model; the direction of the hemispheric asymmetry was found to depend on the type of materials used, whereas its magnitude was determined by the type of memory process. For verbal short-term memory, the HERA model seems to be confirmed. The pre-existing representations of the material could take into account the similarity of the hemispheric asymmetry pattern between short-term memory and long-term memory. In contrast, for visuospatial short-term memory, Baddeley's working memory model seems to better explain our results insofar as the asymmetries were essentially linked to the material in encoding but not in retrieval. This latter difference between short-term memory and long-term indicates that processes involved in LTM depend on episodic processes per se, hence, lending more support for the HERA model. Accordingly, these two memory systems seem to bring into play two different modes of hemisphere specialization.