2.2.1. Participants
Nineteen volunteers (13 females) took part. According to self-report, all were right- handed native-English speakers with normal or corrected to normal vision and no neurological and/or psychiatric disorders. One participant (female) was removed from the analyses due to a level of performance (p(hit)-p(false alarm) = 0.19 for faces) that was significantly poorer than the overall group (z = -2.09). The mean age of the remaining 18 participants was 22.9 years (range = 19-30). All participants gave written informed consent prior to the experiment and were paid £ 10/hour for their participation. The experiment received ethical approval from the Cardiff University School of Psychology Ethics Committee.
2.2.2. Materials
Stimuli comprised a set of 222 black and white images of Caucasian male and female faces with neutral expressions and a set of 222 black and white images of landscapes, streets and buildings devoid of people (Fig. 2.2). These items overlapped with the stimuli used by Taylor et al. (2007) to investigate recognition memory for faces and scenes in patients with MTL lesions. Any scenes that contained distinctive and/or numerous objects were removed, replacing them with additional photographs of landscapes that were taken specifically for this experiment. Additional face stimuli were obtained from a database held at Cardiff University. Twelve items from each set were used in a practice session, with the remaining items separated into 120 faces and 120 scenes presented at study and test (‘old’ stimuli) and 90 faces and 90 scenes presented at test only (‘new’ stimuli).
2.2.3. Tasks and procedure
The experiment was run using E-Prime version 2.0. Images were projected from a stimulus presentation machine to the screen within the scanner, which was manually adjusted for each participant to ensure the image was centred correctly. The MR projector system comprised a Canon SX60 LCOS projector, coupled to a Navitar SST300 zoom converter lens. Two MR compatible button boxes, one for each hand, were employed. The fMRI data were collected on a General Electric 3-T HDx MRI system using an 8 channel receiver-only head coil.
Participants were scanned during study and test phases (Fig 2.2). The experiment consisted o f 3 separate study and test runs, with an equal number of faces and scenes shown in each run. Forty faces and scenes were presented in each study phase; these were seen again in the test phase, which followed immediately, along with 60 unstudied stimuli (30 faces and 30 scenes). No stimuli were encountered in more than one study-test run, and the order of face/scene presentation at study, as well as test, was randomised for each participant-.
Study items were presented in the centre of the screen against a black background for 2000ms, separated by a jittered inter-stimulus interval (ISI: 800ms-1500ms, mean = 1000ms) during which the screen remained black. Participants made a
pleasant/unpleasant judgement for each stimulus, using left and right index fingers. The hand used to signal pleasant/unpleasant judgements was counterbalanced across participants and prompts for the pleasant/unpleasant judgement appeared beneath each study item, which remained onscreen for the duration of each trial.
Test items were presented in the centre of the screen for 4000ms, separated by a jittered inter-stimulus interval (800ms-1500ms, mean = 1000ms) during which the screen was black. Participants were told that they were to distinguish between old (studied) and new stimuli and to indicate their confidence in the old/new judgment on a 6-point scale (1: high confidence new, 2: medium confidence new, 3: low confidence new, 4: low confidence old, 5: medium confidence old, 6: high confidence old; see Fig. 2.2). As with previous experiments that have used a 6-point confidence scale, participants were encouraged to make use o f the entire scale (Yonelinas et al., 2005). Responses were made using the right/left ring (high confidence), middle (medium confidence) and index (low confidence) fingers. The hand used for old judgments was counterbalanced across participants. Throughout each trial, prompts for the high, medium and low old/new confidence ratings appeared beneath the test item. The presentation order of the 3 study/test runs was counterbalanced across participants. E n c o d in g R e trie v a l ( s c a n n e d ) ( s c a n n e d ) 4000ms 2000ms P le asa n t o r U n p leasan t? 8 0 0 - 1500ms 80 0 - 1500ms 4000ms 2000ms P leasan t or
Figure 2.2: Stim uli and procedure. During encoding participan ts m ade pleasant and unpleasant ju dgem ents about novel fa ce s and scenes. A t retrieval participan ts were presen ted with previously studied an d novel fa ce s an d scenes. Participants m ade m em ory ju dgem ents to stimuli at retrieval using a 6-point scale (1 = high confidence new, 2 = m edium confidence new, 3 = low confidence new, 4 = low confidence old, 5 = medium confidence old, an d 6 = high confidence old).
Prior to entering the MRI suite, participants carried out a practice task. They saw 6 faces and 6 scenes at study, and 12 faces and 12 scenes at test (an equal number of old and new faces and scenes). They were also asked to explain the reason for each of their responses at test, to ensure they understood the task and the confidence scale.
2.2.4. Scanning parameters
For functional imaging, a gradient-echo, echo-planar imaging (EP1) sequence was used. The same scanning protocol was used in all runs. Forty-five slices were collected per image volume covering the whole-brain. Scanning parameters were: repetition time/echo time (TR/TE) 2750ms/35ms; flip angle (FA) 90°; slice thickness 2.4mm (3.4 * 3.4 * 2.4mm voxel) with a 1mm inter-slice gap; data acquisition matrix GE-EPI 64 * 64; field of view (FOV) 220 * 220mm; and ASSET (acceleration factor). In addition to this, the first frames were dropped to allow for signal equilibrium. Slices were acquired with a 30° oblique axial tilt relative to the anterior-posterior commissure line (posterior downward). To correct for geometrical distortions in the EPI data due to magnetic-field in-homogeneity, a map of the magnetic field was produced from two 3D SPGR images acquired during the scanning session (Jezzard & Balaban, 1995). The SPGR acquisitions were prescribed using the same slice orientation as the EPI data to be unwarped. Parameters for the SPGR acquisitions were: TE 7ms and 9ms; TR 20ms; FA 10°; data acquisition matrix 128 * 64 * 70; FOV 384 * 192 * 210mm. Anatomical images were acquired using a standard Tl- weighted sequence comprising 178 axial slices (3D FSPGR). Scanning parameters were: FA 20°; data acquisition matrix 256 * 256 * 176; FOV 256 * 256 * 176mm, and
1mm isotropic resolution.
2.2.5. JMRI data pre-processing
This was carried out using FEAT (FMRI Expert Analysis Tool) Version 5.63, which is part of FSL (FMRIB Software Library, www.finrib.ox.ac.uk/fslT The following pre statistics processing was applied; motion correction using MCFLIRT (Jenkinson, Bannister, Brady, & Smith, 2002); non-brain removal using BET (Smith, 2002); spatial smoothing using a Gaussian kernel of FWHM 5mm; mean-based intensity normalisation of the entire 4D data set by the same multiplicative factor; high pass temporal filtering (Gaussian-weighted least-squares straight line fitting, with sigma = 50s). Phase information from the two SPGR images was unwarped using PRELUDE
(Jenkinson, 2003). The unwarped phase images were then subtracted and the resulting fieldmap was used to unwarp the EPI data using FUGUE (Jenkinson, 2003). Time-series statistical analysis was carried out using FILM with local autocorrelation correction (Woolrich, Ripley, Brady, & Smith, 2001). Registration to high resolution 3D anatomical T1 scans (per participant) and to a standard Montreal Neurological Institute (MNI-152) template image (for group average) was carried out using FLIRT (Jenkinson et al., 2002; Jenkinson & Smith, 2001).