[P4–293]: STRATEGY CLASSIFICATION OF CLOCK‐DRAWING TEST CONSTRUCTION

der 40 years old) and 23 older participants (over 55 years old). The study was approved by the Research Ethics Committee of the Faculty of Health Sciences in Trinity College Dublin. All participants signed informed consent before participating in study. The DMS task had 3 trial types: face-name, face and a control (scrambled images). Trials consisted of an encode phase (2s), a delay (18s) and a probe phase (4s), with a jittered inter-trial interval (6-14s). There were 75 trials (25 face-name pairs, 25 face and 25 control stimuli). The regressors for the analysis included in the first case regressors for the encoding, delay and probe periods but these were expanded to subdivide the delay period into three 6 sec period: initial, middle, and late periods. Statistical significance was set at voxel-wise threshold of p 0.005. To correct for multiple comparisons, each cluster had a minimum size of 683ml (23 voxels) which was calculated using AlphaSim (AFNI) to get a family wise error corrected statistical significance level of p<0.05.Results:For the young group the statistical parametric map (SPM) of the face vs control are shown in Figure 1 and with the delay period sub-divided into 3 parts in Figure 2. Similarly the face-name DMTS results are shown in Figure 3 and Figure 4 (split delay periods). In Figure 5, face vs control activation patterns seen in the older healthy groups (split delay) are seen, and Figure 6 contain the results from the face-name vs control task. The statistical maps show that during the delay period with either face or face-name stimuli, the activation pattern varied among the 3 different periods in both young and older cohorts. Conclusions:This study is supportive of the idea that brain activation is not static during a specific cognitive process, but may be dynamic.