Abstracts

It is now well established that left (dominant) anterior temporal lobectomy (ATL) may cause verbal memory deficits (Jones-Gotman, 1991; Milner, 1967). However, when the right (non-dominant) temporal lobe is concerned, visuospatial memory deficits are not always clearly observed (Moore [amp] Baker, 1996; Naugle et al., 1993). It has been argued that the absence of a clear double dissociation with regard to this material-specific hypothesis might be due to the instruments used to assess episodic memory functions (Lee et al., 2002). A second hypothesis proposed to account for this phenomenon relies on the familiarity of the material (familiar/non-familiar) to be learned, rather than on the usual verbal/visuospatial distinction. Some attempts have been made to distinguish the role of familiarity from the role of the type of material in episodic learning and memory, though the results were ambiguous (Falk et al., 2002; Redoblado et al., 2003). Thus, the goal of the present study is to test the two following hypotheses: the material-specific and the familiarity hypotheses. 30 temporal lobe epileptic patients (13 RT; 17 LT) who underwent surgery (ATL or amygdalo-hippocampectomy) for intractable seizures, and 18 normal controls (NC), participated in the present study. In order to test the material-specific and the familiarity hypotheses in episodic learning and memory, a fully crossed design was elaborated and consisted of the four following learning and memory recognition tasks: abstract words (familiar verbal), nonsense words (non-familiar verbal), landscape photographs (familiar visuospatial), and abstract designs (non-familiar visuospatial). Free recall was also assessed for the verbal tasks. On recognition tasks, a repeated measures ANOVA 2 (familiar/non-familiar) X 2 (verbal/ visuospatial) X 3 (immediate 1/immediate 2/delayed] showed the following results. Better performances were observed on non-familiar than familiar tasks, as well as on verbal than visuospatial tasks. On verbal tasks, NC and RT did not differ, and both NC and RT performed better than LT. On visuospatial tasks, NC performed better than both LT and RT, which did not differ from each other. NC performed better than both LT and RT on each recognition (1, 2 and delayed), but no difference was observed between both epileptic groups on these moment measures. Repeated measures ANOVA on free recalls of words and nonsense words showed that participants performed better on the word task than on the nonsense word test. Overall, NC performed better than LT and RT, and LT performed worse than RT. Neither the material-specific hypothesis nor the familiarity hypothesis appear sufficient on their own to explain the obtained data. Indeed, both the familiarity and the type of material to be learned play a role in episodic learning and memory. (Supported by Savoy Foundation)