Magnocellular input [3,8], have been shown to contain directionally selective cells [9] or to be responsive to motion in human functional magnetic resonance imaging (fMRI) experiments, though less robustly than V5 [10 ?7].3. Material and methods3.1. Ethics statementInformed written consent was obtained from all participants and the University College London (UCL) Research Ethics Committee approved the study.3.2. SubjectsNineteen healthy subjects (10 males; minimum age 21, maximum age 56, mean age 32) were recruited through advertisements and from the UCL psychology subject pool; three of them were excluded after being scanned because their rating data had not been correctly recorded. None of our participants was an artist and all had normal or corrected-to-normal vision.3.3. StimuliStimuli were generated using PROCESSING (www.processing. org) and then passed to COGENT 2000 and COGENT GRAPHICS (www.vislab.ucl.ac.uk/cogent.php) for playback. Subjects were shown and asked to rate the stimuli twice: once during a visit to the laboratory one or more days before scanning, when the experiment was also explained to them, and once in the scanner. In the pre-scanning session subjects sat in a darkened room at a fixed distance from a cathode ray tube computer display. They rated the patterns using a computer keyboard. The responses in this part of the study were on an 8-level scale. Stimuli Sch66336 web consisted of eight patterns of moving white dots on a black background, designed with the expectation that some patterns would be preferred to others. The patterns were generated algorithmically using trigonometric functions or particle systems [18], and were matched for the number of dots and their linear velocities at the five velocity levels used. In the pre-scanning experiments, subjects sat at a distance of 60 cm from the display and each dot of the display subtended 0.58 of visual angle. In the scanner, subjects were positioned 8 55 cm from the display, which had a width of 31 cm. The actual area used to display the stimuli was adjusted so that each subject was able to see the entire field of dots. Each individual Y-27632 web stimulus contained 192 dots. The speed of the dots’ motion in the pre-scanning sessions was varied in five steps, corresponding to 2.86, 5.73, 8.59, 11.46 and 14.32 deg s21, while speeds in the scanning session varied based on how the stimuli were scaled down for display. The mean dimensions of the area in which the stimuli were shown after individual adjustment were 29 ?238 . A single 8 dot subtended a visual angle of approximately 0.178 , the 8 size reduction being necessary to make the entire stimulus area visible to the subject in the scanner. The stimuli are available for viewing at www.vislab.ucl.ac.uk/kinetic_ beauty_movies. We emphasize that the two patternspreferred by the majority of subjects had different characteristics and both differed in their characteristics from a pattern preferred by one of the other subjects (see below and ?). Two subjects were given two stimulus sessions in the scanner owing to a recording failure (and only data from the second session was used for them); the remaining subjects were given one session. The session began with a 26 s period with no stimulus on the screen. Brain volumes recorded during this period were discarded from subsequent analysis to allow T1 equilibration effects to subside. The stimulus sequence began after this period. A block design was used to plan the timing of the stimuli. The patter.Magnocellular input [3,8], have been shown to contain directionally selective cells [9] or to be responsive to motion in human functional magnetic resonance imaging (fMRI) experiments, though less robustly than V5 [10 ?7].3. Material and methods3.1. Ethics statementInformed written consent was obtained from all participants and the University College London (UCL) Research Ethics Committee approved the study.3.2. SubjectsNineteen healthy subjects (10 males; minimum age 21, maximum age 56, mean age 32) were recruited through advertisements and from the UCL psychology subject pool; three of them were excluded after being scanned because their rating data had not been correctly recorded. None of our participants was an artist and all had normal or corrected-to-normal vision.3.3. StimuliStimuli were generated using PROCESSING (www.processing. org) and then passed to COGENT 2000 and COGENT GRAPHICS (www.vislab.ucl.ac.uk/cogent.php) for playback. Subjects were shown and asked to rate the stimuli twice: once during a visit to the laboratory one or more days before scanning, when the experiment was also explained to them, and once in the scanner. In the pre-scanning session subjects sat in a darkened room at a fixed distance from a cathode ray tube computer display. They rated the patterns using a computer keyboard. The responses in this part of the study were on an 8-level scale. Stimuli consisted of eight patterns of moving white dots on a black background, designed with the expectation that some patterns would be preferred to others. The patterns were generated algorithmically using trigonometric functions or particle systems [18], and were matched for the number of dots and their linear velocities at the five velocity levels used. In the pre-scanning experiments, subjects sat at a distance of 60 cm from the display and each dot of the display subtended 0.58 of visual angle. In the scanner, subjects were positioned 8 55 cm from the display, which had a width of 31 cm. The actual area used to display the stimuli was adjusted so that each subject was able to see the entire field of dots. Each individual stimulus contained 192 dots. The speed of the dots’ motion in the pre-scanning sessions was varied in five steps, corresponding to 2.86, 5.73, 8.59, 11.46 and 14.32 deg s21, while speeds in the scanning session varied based on how the stimuli were scaled down for display. The mean dimensions of the area in which the stimuli were shown after individual adjustment were 29 ?238 . A single 8 dot subtended a visual angle of approximately 0.178 , the 8 size reduction being necessary to make the entire stimulus area visible to the subject in the scanner. The stimuli are available for viewing at www.vislab.ucl.ac.uk/kinetic_ beauty_movies. We emphasize that the two patternspreferred by the majority of subjects had different characteristics and both differed in their characteristics from a pattern preferred by one of the other subjects (see below and ?). Two subjects were given two stimulus sessions in the scanner owing to a recording failure (and only data from the second session was used for them); the remaining subjects were given one session. The session began with a 26 s period with no stimulus on the screen. Brain volumes recorded during this period were discarded from subsequent analysis to allow T1 equilibration effects to subside. The stimulus sequence began after this period. A block design was used to plan the timing of the stimuli. The patter.