No abstract

No abstract

No abstract

The salience of targets in preattentive tasks depends on local feature gradients between the target and the surrounding background. We have developed an electrophysiological method to isolate texture-segregation-specific components in the human visual evoked potential (tsVEPs). We report here on threshold estimates of orientation-based texture segregation, comparing psychophysical and electrophysiological measures. Eight visually normal subjects participated in the experiment. In a 4AFC task, subjects indicated the position of a single patch that differed in orientation from a homogeneously oriented background. Orientation gradients ranged from 5° to 90°; orientation threshold was estimated with a best PEST procedure. In further experiments VEPs to the onset of ‘orientation chequerboard’ patterns with orientation gradients ranging from 0° to 90° were recorded from an Oz-Fpz derivation. Orientation threshold was estimated in two steps. First, the VEP to the 0° condition was subtracted from all other VEPs to isolate tsVEPs. Second, tsVEP amplitudes were determined by cross-correlation between the tsVEP to any given orientation and the 90° condition. We found, first, that the electrophysiological threshold derived from the overall mean tsVEPs was considerably higher than the psychophysical threshold of the orientation gradient, 16.1°±1.0°. Thus the pool of neurons contributing to the VEP signal must be larger or more strongly activated in order to elicit a significant tsVEP than to subserve psychophysical performance. Second, we found that the amplitudes of tsVEPs above threshold increased monotonically with increasing orientation gradient of the chequerboard patterns, suggesting that the tsVEP may have its neural correlate in orientation-contrast-dependent neurons similar to those found in monkey and cat striate cortex.

The “sequential neural network” proposed by Jordan (MIT COINS Technical Report 88-27, 1988) enables learning of motor skill problems involving excess degrees of freedom. This structure was used with (i) the articulatory model developed by Meada [J. Acoust. Soc. Am. Suppl. 1 65, S22 (1979)] as an internal “forward” model relating a given set of articulatory commands (five articulators, e.g., lips, jaws, tongue body, tongue dorsum, and tongue tip) to their acoustic consequences, (ii) vocalic targets specified in the formant (F1, F2, F3) space, and (iii) smoothness constraints leading to coarticulation phenomena. A first set of results about learning of such vocalic gestures will be presented and the following will be discussed: (i) the pattern of coarticulation phenomena observed in this structure, such as between-articulator relationships for achieving a given formant target and temporal organization of the trajectories in the articulatory space, (ii) the role of gesture duration, and (iii) the compensation ability of the network in relation to perturbations such as those observed in the bite-block experiments by Lindblom et al. [J. Phonet. 7, 147–162 (1979)].