E. S. Mikhailova

Gender differences in the recognition of spatially transformed figures: Behavioral data and event-related potentials (ERPs)

The gender differences in accuracy, reaction time (RT) and amplitude of the early P1 and N1 components of ERPs during recognition of previously memorized objects after their spatial transformation were examined. We used three levels of the spatial transformation: a displacement of object details in radial direction, and a displacement in combination with rotation of the details by ±0° to 45° and ±45° to 90°. The accuracy and the RT data showed a similarity of task performance in males and females. The effect of rotation was significantly greater than the effect of simple displacement, and the accuracy decreased, and the RT increased with the rotation angle in both genders. At the same time we found significant sex differences in the early stage of visual processing. In males the P1 peak amplitude at the P3/P4 sites increased significantly during the recognition of spatially transformed objects, and the wider the angle of rotation the greater the P1 peak amplitude. In contrast, in females the P1 peak amplitude did not depend on the rotation of figure details. The N1 amplitude revealed no gender differences, although the object transformation evoked somewhat greater changes in the N1 at the O1/O2 sites in females compared to males. This new fact that only males demonstrated the sensitivity of early perceptual stage to the transformation of objects adds information about the neurobiological basis of different strategies in the visual processing used by each gender.

Location of the dipoles of the P1 wave of visual evoked potential in the human brain

Line fragments and their intersections are the basic properties of most objects in the visual world [3]; therefore, their rapid and reliable detection is important for image recognition. It is obvious that identification of lines (properties of the first order) and their intersections (properties of the second order) in images occurs in cats and monkeys already in the primary visual cortex where half of neurons detect the orientation of single lines and the other half of neurons more rapidly and strongly respond to their intersections or branching nodes [11, 12]. There is no information on the location of the mechanism responsible for identifying lines and their intersections in the human brain. Our previous study [1] demonstrated differences in the temporal and amplitude characteristics and regional specificity of the components of visual evoked potentials (EPs) in the human brain after presentation of images consisting of lines and crosses. However, the location of the dipole sources of these EP components has not been studied. Nevertheless, such information is important for assessing more precisely the involvement of different zones of the visual cortex in the identification of image properties. Therefore, we studied threedimensional location of dipole sources of the early component of visual EPs (wave P1) in the spherical and real model of the head in 12 subjects during presentation of images consisting of lines and crosses. The P1 component was selected because, first, it is associated with early stages of the processing of simple image features in the visual cortex, and, second, according to fMRT data, the location of the current dipole of the P1 wave coincides with the activation focus in the visual cortex [4]. METHODS The EEG was recorded in the electrophysiological experiments on 12 healthy subjects using a NeocortexPro 40-channel recording system (Neurobotics, Russia) according to the 10‐10 scheme with a sampling rate of 1000 Hz/channel. The right-ear electrode was used as an indifferent electrode. The upper limit of the frequency band of amplifiers was 100 Hz, the lower limit was 0.1 Hz, and the characteristic slope was 12 dB/octave. The subjects studied sat in a chamber isolated from light with a background illumination of 6 cd/m 2 . The images containing sets of lines or crosslike figures were shown for 100 ms on a monitor in a random order with various intervals [1]. The angular size of the image was 18.8° , and the size of its single element (cross or line) was 0.6° . Each experiment included 50 stimuli of each type. The average optical densities of the stimuli were equalized. A short sound warning was given 1 s before the beginning of the stimulus exposition. The subject focused the gaze at the fixation point in the screen center after the signal and watched the appearing image.

Trajectories of shifting dipole sources of visual evoked potentials across the human brain

Visual evoked potentials in response to images of a set of horizontal and vertical lines or crosses were recorded from the brains of 18 human subjects in 34 leads. Inverse EEG analyses were used for the dynamic location of the dipole current sources of the N1, P1, and N2 waves using a two-dipole spherical model with a 1-msec step. The occipital lobes of all subjects showed significant displacement of the dipoles of evoked potential waves along predominantly arc-shaped trajectories (75.8% of cases). Trajectory durations (average about 25 msec) were characterized by insignificant interindividual variability and were independent of the type of stimulus and the phase of the evoked potential. A characteristic (occurring in 85% of cases) “jump” in the coordinates of the dipole, which constituted a rapid, sharp, and significant medial displacement, was seen between the first and second trajectories of the equivalent current dipoles (at 110–120 msec after stimulus onset). The possible significance of these data for understanding the dynamics and kinetics of processing of local image features in the human visual cortex is discussed.

Behavioral and electrographical features of recognition of forward-masked complex images: The effect of categorical similarity of the target and masking stimuli

In psychophysical and neurophysiological experiments, the subjects recognized images of two categories, “animals” and “objects.” The images of the same categories differing from the target stimuli were used as the masking stimuli. We found that the efficacy of forward-masking depended on categorical similarity of the masking and target stimuli. The probability of a correct response was lower and the reaction time and its variance were higher when we used stimuli of the same category compared with presentation of stimuli of different categories. Categorical similarity of masking and target stimuli induced difficulties with response, which were accompanied by decreases in the amplitudes of the N2 and P3 components of the evoked potentials. These effects were more pronounced during recognition of animals compared to objects. The results are discussed from the point of view of negative priming and the distractor effect.

Temporal and topographic characteristics of evoked potentials in the conflict of two consecutive visual stimuli in a working memory task

Behavioral reactions and brain mechanisms involved in processing two matching or mismatching (conflicting) visual stimuli were studied in healthy subjects (mean age 22.57 ± 0.46 years). Line orientations (vertical, horizontal, or 45°) were used as stimuli and were presented with an interval of 1500–1800 ms. The reaction time was shown to increase in the case of a conflict of two orientations as compared with matching orientations. The reaction time depended on the orientation of the reference stimulus and was minimal when a vertical line was used as a reference. An increase in N2 negativity (time window 200–280 ms) in the frontal and parietal cortical areas was identified as an informative indicator of a conflict between the current orientation and the orientation stored in working memory. The dipole sources of N2 were localized to the prefrontal cortex (middle frontal gyrus, frontal pole, and pars orbitalis). The N2 amplitude was found to depend on the orientation of the first stimulus in a pair, being higher in the case of a 45° orientation. The visual areas were shown to play a role in detecting a conflict of two consecutive signals because the early sensory components increased in amplitude. The results implicate cortical structures, including the sensory-specific visual, parietal, and prefrontal areas, in comparing consecutive visual signals and detecting their conflict.

Latency of evoked potentials in the tasks involving classification of images after wavelet filtration

We studied the characteristics of evoked potentials recorded during the recognition test based on four types of series of images subjected to the wavelet filtration: images of living objects containing either low frequency or high frequency portion of the spatial frequency spectrum, and imaging of non-living objects in the same two spatial frequency bands. Each subject had to classify the image either by its semantic feature (living–non-living), or by its physical feature (low frequency–high frequency). The purpose of this study was to compare the time characteristics of evoked potentials in these two types of tasks, which provides information on the time characteristics of categorization mechanisms of visual images. Analysis of the latent periods and amplitudes of the components of evoked potentials allowed us to detect the occipital areas of the leads where the early components (up to 170 ms) are associated with spatial and frequency characteristics of the image, the frontal and temporal areas where the components of 170–200 ms correspond to the process of categorization, and the later frontal, central, and parietal areas (300–500 ms) correspond to the process of error detection and the organization of motor response.

Mechanisms of orientation sensitivity of human visual system: Part II. Neural patterns of early processing of information about line orientation

A high-density EEG was recorded in 41 healthy subjects (20 men and 21 women) in the cardinal (horizontal and vertical) and oblique (45 and 135 degrees) line identification task. The analysis of the adaptive amplitude maximum/minimum (4 ms averaging) of P1 and N1 components of evoked potentials in the symmetrical occipital, parietal, and inferior temporal areas and dipole source modeling showed anisotropy of cortical responses in the 80- to 150-ms interval. The amplitude is higher for the oblique orientations as compared with cardinal ones. The temporal and regional features of cortical responses were revealed. The earlier selective response (∼90 ms latency) is recorded in the parietal areas, while the later response (∼140 ms latency) is found in the occipital areas. A number of sex-related differences were discovered at the early stages of line orientation detection: in men, the amplitude of the components is higher; they have a broader area of localization of their dipole sources; in addition to the occipital and parietal regions, the inferior temporal cortex is also involved. The data obtained are discussed in the context of the notion of effective neural coding and the features of spatial information processing in the visual system of men and women.

Gender-Related Strategies for Solving Visuospatial Tasks

Studies of 34 subjects (of which 16 were men) using a gender-related differences model addressed the mechanisms forming two strategies for solution of a visuospatial construction task. While there were no gender-related differences in the effectiveness of performing the construction task, the patterns of evoked activity in men and women were different. In men, the early response in the parietal cortex was linked with spatial transformation of the figure: the greater the rotation of the constituent parts, the greater the amplitude of the P1 wave, while errors were associated with decreases in P1 amplitude. In women, no cortical correlates of the rotation of parts were seen, though there was an increase in the negativity of N150 in EP in the occipital and inferior temporal areas of the cortex on transformation of the whole figure into the set of its component parts. These data are assessed in the light of concepts of the gender specificity of representations of the visual space and the cerebral organization of different strategies of visuospatial activity.

Evoked Visual Cortical Potentials in Humans during Perception of Whole Figures and Their Constituent Elements

Changes in evoked potentials were studied in 32 subjects during perception of whole and disintegrated images. In the occipital and parietal areas, responses to disintegration arose early, during the development of P1 waves, and their characteristics were defined by the magnitude of the response to the whole figure. In the occipital cortex, subjects of group 1 had low-amplitude P1 waves, which increased on image disintegration, while the high-amplitude P1 waves in subjects of group 2 showed a tendency to decrease. In the parietal areas, effects were clear only in subjects of group 1 and were different in the right and left hemispheres: on the left, P1 amplitude increased with the appearance of simpler elements in the image, while on the right, changes in the spatial disposition of details were more significant. In the inferior temporal cortex, responses to disintegration consisted of decreases in the amplitude of the later N1 wave, which were significant only in subjects of group 2. The appearance of simpler elements in images led to increases in the P3 wave in all groups. These data demonstrate topographical and temporal specificity in the responses of the visual cortex to image disintegration and the possible existence of different strategies at the early stages of visual image analysis.

Studying the dynamics of visual perception using a dipole model

A study of the encoding of the basic attributes of an image in the human visual cortex by means of moving dipoles has shown for the first time that, in the 50–300-ms interval after the stimulus, equivalent current dipoles of induced-potential waves lasting about 27 ms are displaced predominantly along curved trajectories. At 110–120 ms from the beginning of the stimulus, there is an abrupt displacement of the dipole from a lateral to a medial position. Two zones of preferred localization of the dipoles are detected in the lateral and medial regions of the visual cortex, whose coordinates coincide with the beginning and end of the trajectories, while the size varies as a function of the phase of the potential. The resulting data are important for estimating the dynamics and kinetics of the processing of the attributes of an image in the visual cortex of the human brain.