Search

When two masked targets (T1 and T2) are visually or auditorily presented in rapid succession, processing of T1 produces an attentional blink (AB)--that is, a transient impairment of T2 identification. The present study was conducted to compare the relative impact of masking T1 and T2 between vision and audition. Within a rapidly presented sequence, each of the two verbal targets, discriminated by their offset (Experiment 1) or their onset (Experiment 2), could be followed by either a single item, acting as a mask, or a blank gap. Masking of T2 appeared to be necessary for the occurrence of the AB for both the visual and the auditory modality. However, whereas masking of T1 affected the expression of the visual AB in both experiments, the same effect was observed in the auditory modality only when the targets varied at the onset. These results provide further evidence that processing auditory and visual information is restricted by similar attentional limitations but also suggest that these limits are constrained by properties specific to each sensory system.

Much work has been done to investigate participants’ ability to detect repeated targets, but the present study examined the influence of recently attended stimuli on target masking. Participants performed a target identification task in which the item that masked the target was either a recently attended item or a novel item. When it was identical to a previously attended stimulus, the mask was rendered considerably less effective. We have termed this effect a repeated mask reduction (RMR). This simple manipulation resulted in a large, reliable effect on the efficacy of visual masking in a single-target identification paradigm. In Experiments 1, 2, and 3, we demonstrated the basic effect and noted that the RMR increased as task difficulty increased. Experiments 4 and 5 suggested that attention to the first instance of the mask was crucial to this effect by showing that the magnitude of the RMR was unaffected by repetition of nonmask distractors and that the magnitude of the effect was reduced when less attention was paid to the first instance of the mask.

A briefly presented visual target stimulus can be difficult to identify if shown in the context of a mask stimulus. There are several quantitative models that have been used to explain many different masking effects. Here, we look at modeling what has been called object substitution. We analyze four models and show that three of the models work with a common principle, whereas the fourth behaves quite differently. The three similar models have previously been used to explain many other aspects of visual masking, whereas the fourth model was designed exclusively to deal with object substitution masking effects. We identify an experimental test on the basis of which to choose between the models. The experimental results support the behavior of the fourth model.

When attention is divided, a briefly presented target surrounded by four small dots is difficult to identify when the dots persist beyond target offset, but not when these dots terminate with the target. This object-substitution masking effect likely reflects processes at both the image level and the object level. At the image level, visual contours of the mask make feature extraction difficult. Recent data (Lleras & Moore, 2003) suggest that, at the object level, an object file is created for the target-plus-mask, and this single-object token later morphs into a single-object token containing the mask alone. In the present experiments, we used stimuli presented in 3-D space and apparent motion; the results indicate that object-substitution masking also arises when the mask and the target are represented in two separate object tokens and the mask token interferes with the target token.

Motor responses can be facilitated by congruent visual stimuli and prolonged by incongruent visual stimuli that are made invisible by masking (direct motor priming). Recent studies on direct motor priming showed a reversal of these priming effects when a three-stimulus paradigm was used in which a prime was followed by a mask and a target stimulus was presented after a delay. A similar three-stimulus paradigm on nonmotor priming, however, showed no reversal of priming effects when the mask was used as a cue for processing of the following target stimulus (cue priming). Experiment 1 showed that the time interval between mask and target is crucial for the reversal of priming. Therefore, the time interval between mask and target was varied in three experiments to see whether cue priming is also subject to inhibition at a certain time interval. Cues indicated (1) the stimulus modality of the target stimulus, (2) the task to be performed on a multidimensional auditory stimulus, or (3) part of the motor response. Whereas direct motor priming showed the reversal of priming about 100 msec after mask presentation, cue priming effects simply decayed during the 300 msec after mask presentation. These findings provide boundary conditions for accounts of inverse priming effects.

Does unconscious perception at the objective detection threshold (ODT) exist? In this reply, formal meta-analyses of previously narratively reviewed evidence (Snodgrass, Bernat, & Shevrin, 2004) provide strong evidence for large, reliable unconscious perceptual effects, and further strengthen previous conclusions for their ODT status. These meta-analyses dispel general concerns that ODT effects are small, unreliable, or susceptible to file drawer concerns (cf. Reingold, 2004). Furthermore, the specific objections (Holender & Duscherer, 2004) to individual ODT effects are refuted, and objections to our proposed solutions to the exhaustiveness, exclusiveness, and null sensitivity problems (Haase & Fisk, 2004; Reingold, 2004) are shown to be inapplicable, contradicted by the available evidence, or both. Accordingly, the strong evidence for ODT effects, taken together with the nonmonotonic relationship, contradicts the single-process conscious perception model (e.g., Holender & Duscherer, 2004) and provides stronger evidence for unconscious perception than hitherto available.

When two targets are displayed in rapid visual sequence and masked by trailing patterns, identification accuracy is nearly perfect for the first target but follows a U-shaped pattern over temporal lag for the second target. Three experiments examined the role of visual masking in this attentional blink. Experiment 1 compared integration and interruption masks for both targets. Although either mask was effective in producing the blink when applied to the first target, only the interruption mask was effective when applied to the second target. Experiment 2 showed that integration masking of the second target was ineffective over a wide range of accuracy levels. Combining the two forms of masking in Experiment 3 confirmed the dissociation: A combined mask and only a main effect on accuracy for the first target, whereas it produced a qualitatively different pattern over temporal lag for the second target. These results suggest that representations of the target are substituted in consciousness by that of the interruption mask when visual attention is preoccupied.

The effect of a pretarget (TS1) masked by a noise mask (MS) upon the recognition of a posttarget (TS2) was examined in a TS1-MS-TS2 paradigm. The results clearly showed that the presentation of TS1 masked completely by MS reduced the recognition of a subsequent TS2 to a much greater extent if TS2 was the same as TS1 than when it was not. The results suggest that a TS1 that is perceptually masked by an MS nevertheless continues to have an effect within the visual system. A possible explanation of this reductive effect is described briefly.

A pattern made by randomly filling cells in a square matrix was presented for 1 see and followed, after various intervals, by an identical or similar pattern. Ss responded “same” or “different.” Performance was fast and accurate if the interval was short and there was no movement or masking of the pattern during the interval. Performance was slower, less accurate, and highly dependent on pattern complexity if the interval exceeded 100 msec or if there was movement or masking. The results are interpreted as evidence for two distinct classes of visual memory: high-capacity sensory storage which is tied to spatial position and is maskable and brief; and schematic short-term visual memory which is not tied to spatial position, which is protected against masking, and which becomes less effective over the first few seconds but not over the first 600 msec.

In duration, size, and form discrimination tasks, a visual noise mask was presented at variable delays after stimulus offset in order to interrupt processing and control the extent of processing time. Previous work (Thomas & Cantor, 1975) had suggested that both perceived duration and perceived nontemporal “information” might be expected to increase as processing time was extended. As predicted, accuracy in the discrimination of size of circles and form of non-sense figures was found to vary directly with stimulus duration (20, 50 msec) and mask delay interval (0, 30, 70, 110 msec). Differences in perceived duration between filled (forms or circles) and unfilled (blank) intervals were found to increase monotonically with increases in the mask delay interval, when non-sense forms, but not circles, were presented. Two hypotheses of visual masking (“integration” and “interruption”) are discussed. Within the context of the “integration” hypothesis, a model is proposed which predicts processing time as a function of stimulus duration, mask delay interval, and the interval between onset of the mask and termination of processing.

The stimuli consisted of two complementary dot patterns that formed a bigram when they were flashed simultaneously; impairment of letter recognition developed when one of the patterns was briefly extended beyond the termination of the other (stimulus offset asynchrony). However, if the ratio of stimulus offset asynchrony to bigram duration remained constant, the probability of a correct recognition response also remained constant as duration varied over a 50- to 100-msec interval. When percent stimulus asynchrony increased, the impairment increased. An interaction between bigram letter position and each of bigram duration and percent stimulus asynchrony was observed with recognition accuracy greater in general for the letter in the left half of the field.

Visual masking between nonoverlapping gratings depends on the similarity of their spatial frequencies. In a metacontrast experiment, bar gratings of 6 and 12 cycles/deg were masked by flanking gratings of various spatial frequencies. For three observers, masking was greatest when the masking frequency was approximately equal to the masked frequency. The results are interpreted in terms of a theory of reciprocal inhibition among frequency-specific neural units.

A model for visual masking based on the notion of interaction of the specific and nonspecific afferent visual systems is presented. Discussion is focused on nonmonotonic masking functions. In particular, it is proposed that in order for visual information (patterns, forms, etc.) to be consciously perceived, both specific retina-genicula-striate impulses and nonspecific retina-reticulacortical impulses should converge in the same cortical space. Nonspecific activity is shown to be necessary for subjective awareness. This activity is shown to be of longer latency than specific activity. It is concluded that trailing conscious-experience-generating impulses are elicited by collateral activity from the specific information received from thefirst stimulus. These impulses reach the cortex at the same moment as the specific activity of thesecond (masking) stimulus as coded, which has a relatively higher signal-to-noise ratio in the given retinotopically specified cortical space. Consequently, subjects consciously perceive the second stimulus. This operation of awareness generation is termedperceptual retouch and is considered as a special psychological mechanism worthy of psychophysical study.

Visual masking effects on test flash thresholds were measured under real and simulated eye movement conditions to determine whether visual masking is primarily responsible for elevations in threshold that are sometimes associated with saccadic eye movements. Brief luminous flashes presented to the central retina before, during, and after saccades were masked by stimuli presented either pre- or postsaccadically. The amount and time course of masking were quantitatively dependent on stimulus parameters of intensity and temporal separation and were unaffected by eye movement parameters (amplitude, velocity, direction) as long as retinal stimulus conditions were constant. The duration of forward masking was longer than that of backward masking. When retinal conditions during saccades were mimicked while the eyes were held steady, masking interactions were identical to those obtained during real saccades. These results indicate that masking effects during saccades in ordinary environments are determined solely by the stimulus situation at the retina. Putative nonvisual, centrally originating saccadic suppression suggested by other authors is evidently not additive with visually determined masking during saccades.

For nonmetacontrast (disk-disk) masking, if the outer diameter of the mask equals that of an annulus which produces typical metacontrast functions, U- and W-shaped functions will be obtained if (1) the stimuli are viewed dichoptically and monoptically, respectively. and (2) the mask is no more than about 10 times the energy (luminance x duration) of the target. As this energy ratio is approached, both types of functions become monotonic. These shapes suggest that interactions at different visual-pathway loci may he mapped by visual masking functions.

Ss were briefly shown pairs of letters under backward masking and disinhibitive backward masking conditions. For the backward masking condition, a single homogeneous flash of light (M1), varying in duration, followed the presentation of letters after a varying delay (IS11), and recognition of the target letters (TS) was found to be a function of M1 duration and ISI1. For the disinhibition condition, a variable second mask (M2) followed M1 after a variable delay (ISI2) and disinhibition of M1’s effect on TS recognition occurred. Provided the ISIs were adequate, a relatively powerful M2 wass found to disinhibit a moderate Ml’s effect on the TS. Current theories of visual masking were examined in the light of these findings.

Spatial complements of visual displays of one- and two-dimensional Markov spatial constraints were successively presented for discrimination. Although each complement alone was sufficient for discrimination, spatial complements interfere when presented successively in time. Substantial changes in performance are observed for temporal separations as short as 10 msec and for spatial separations as short as 1 min in visual angle. Complete processing appears to take place in the interval between successive displays: thresholds are constant for a constant sum of display duration plus the interfield interval. Implications of complementary interference for a general theory of visual masking are briefly discussed.

In this paper, consideration is given to the physiological foundations of visual masking with successive stimuli. Three experiments are performed with dot patterns. The results indicate that none of the classic characteristics of lateral inhibitory interaction are obtained with this type of stimulus material. A rather informal demonstration is then described that suggests further that lateral inhibitory interaction may also not be a satisfactory explanation of metacontrast and other kinds of masking that use continuous figures.

Forward and backward visual masking for patterns was investigated as a function of relative overlap between test and masking stimuli, relative intensity of masking stimulus and length of interval between stimuli. The extent of masking increased with increasing spatial overlap between stimuli and with increasing intensity of the masking stimulus. Increasing the interval between the stimuli decreased masking; this occurred at faster rates in backward than in forward masking. Possible mechanisms explaining these findings are discussed.

The recognition of alphabetic characters composed of dot patterns is interfered with by dynamic visual noise (DVN) composed of similar dots randomly plotted over the character field. In this study we examine the general ability of the visual system to defect test characters as the level of this DVN varies. Three experiments are performed. The first deals with the effect of the DVN when the characters are embedded in continuous bursts of the noise, while the second and third examine the effects of leading and trading bursts of the DVN on character recognizability. Though the applicability of this paradigm to psychological questions is very brood, we concentrate on the problems of the channel capacity of the visual system and the persistence of the effects. The time course of the psychological instant is also considered.

Two experiments are reported which investigated whether or not the offset of light can serve as a backward masking stimulus (MS). In both studies, human Os made identification responses to graphemes (TS) presented tachistoscopically on a lighted field. Simultaneously with or at short intervals after TS offset, the lighted background field was shut off. The termination of the background field served as the MS. The results indicated a reliable masking effect due to light

We report the results of a study that investigated the effects of four experimental parameters on the tachistoscopic detectability of dotted lines in a noisy background consisting of dynamic patterns of identical dots. Characteristics of the lines that were explored included dot spacing, dot numerosity, and line orientation. Dot spacing was shown to be the most powerful parameter in determining the masking effect: the closer the dots, the more easily the line was detected. Increasing the number of the dots was also effective in increasing the detectability of a line, but only up to a limit of five dots. The orientation of the line, on the other hand, did not alter its detectability. Increasing the density of the visual masking noise monotonically reduced the detectability of the line. These studies are an extension of French’s (1953, 1954) work, differing mainly in the fact that the stimulus material was presented tachistoscopically rather than in a mode that allowed prolonged (up to 5 sec) examination of the figures. Thus, they put the results in a domain in which physiological interpretations may be more critically evaluated. A comparison of the results suggests that essentially the same sorts of cognitive mechanisms are operating in the millisecond domain as in the longer exposures. This suggests that simple neurological models utilizing the concept of feature filters or detectors may be being prematurely applied to relatively complex perceptual processes.

The usual increase in reaction time which occurs when a reaction stimulus is preceded by one irrelevant stimulus was-reduced 30% by preceding the latter with a second irrelevant stimulus at certain critical intervals between stimuli. The interaction between the irrelevant stimuli, as measured by reaction time, has a different time course than the interaction implicit in tlte “psychological refractory period”.

Two separate experiments were conducted to study the visual evoked potential (VEP) correlates of sequential blanking, a phenomenon in which up to one-half of a discrete train of stimulus inputs are not perceived for certain orders of input. A PDP-7 digital computer was used to present stimuli (from 2 to 5 letter Xs) on a CRT display in both experiments. In Experiment 1 there were four basic conditions designed and counterbalanced to indicate the nature of the VEP when stimuli were blanked and when Ss reported all stimuli. All of the stimuli were of equal intensity. The main finding was that although the eight Ss did not perceive and report blanked stimuli, they did respond to them physiologically as indicated by the VEP. The implications of these findings were discussed in relation to recent studies of visual masking and metacontrast in which VEP was recorded. In Experiment 2 the normally blanked and the normally blanking stimuli were alternately increased in intensity to determine the effect on sequential blanking and the VEP. Six Ss were tested under three basic counterbalanced conditions. It was found that sequential blanking could be reliably overcome by increasing the intensity of the normally blanked stimuli. In addition, when the normally blanking stimuli were of greater intensity than the blanked stimuli, not only did perceptual suppression occur, but the evidence indicated that there was no VEP to the first of the two blanked stimuli.

The perception of closely placed visual materials has received considerable attention and has been found to be influenced by a variety of conditions of viewing. Of particular interest to us is the fact that the recognition of symbols in an array is often described as a U-shaped function of distance from a fixation point. In view of increased interest in immediate memory processes, most recent multisymbol recognition work has utilized tachistoscopic (brief display) conditions of viewing (e.g., Estes & Wolford, 1971; Shaw, 1969). In an earlier study in this series (Townsend, Taylor, & Brown, 1971), we were interested in showing that the U-shaped function was not dependent upon time-constrained encoding or processing mechanisms. Accordingly, Ss viewed linear arrays of letters with unlimited viewing time. Recognition of letters was found to be a U'-shaped function of distance from a fixation point. Since the data could not be accounted for in terms of retinal locus effects or memory variables, we concluded that some form of lateral masking effect, the disruption of encoding processes due to adjacent contours, was involved. Since these effects could arise from either retinal or central processes, the studies reported herein were designed to provide locus information concerning this phenomenon.

Two masking experiments were carried out. In the first, duration thresholds were measured for a 10 min black test disc paired with a larger concentric black mask, ranging in size from 15 min to 2 deg. The stimuli were tachistoscopically presented centrally, or at 2 deg or 6 deg in the left binocular field. As mask diameter increased, test threshold decreased in a negatively accelerated function, which approached an asymptote below the unmasked condition. All functions are similar with systematic upward shifts for more peripheral stimulation. In Experiment 2, threshold luminance was adjusted for a 1 deg, 5-msec test flash paired with a 250-mseC., 34-mL mask, ranging from 1 deg to 6.2 deg in diameter. Stimuli were presented in Maxwellian view at 7.2 deg in the right eye nasal field. Results were similar to Experiment 1, except that the asymptote is significantly above the control condition. Both experimental results support a border inhibition hypothesis.

Two experiments were carried out to investigate visual masking and target recovery. In the first experiment it was found that with superimposed disks of various sizes masking increases with increasing target-mask similarity and contour proximity. The results of the second study show that the extent of masking produced by a disk, which is slightly larger than the target, is drastically reduced when the mask in turn is followed by a ring, the inner contour of which is contiguous with the contour of the mask. This “facilitation effect” decreases as the interval between the mask and the third stimulus is increased. Target recovery in this situation appears to be brought about as a result of the reduction or elimination of the contour of the mask, which, as the first study has shown, is critical in bringing about the masking effect. The results are discussed in terms of some of the known physiological properties of the visual system.

Children, 4-5 and 9-10 years old, and college-age adults were tested on two visual masking tasks. The first task required the Ss to identify whether a tachistoscopically presented bar was horizontal or vertical The second task, also using tachistoscopic presentation of stimuli, required Ss to locate a horizontal (or vertical) bar in a matrix of vertical (or horizontal) bars. In both tasks, backward masking produced greater disruption than did forward masking, and the amount of disruption induced by both backward and forward masking decreased as age increased. An Age by Masking Condition interaction was found only in the location task and reflected a much greater difference between backward- and forward-masking conditions for the youngest group than for the older groups. On the basis of these findings and other considerations, it was concluded that only in the location task, which presumably required visual search, was the speed of visual processing slower in the younger group.

Walsh (1971) has added a useful analysis of the possible shape that hypothetical interacting functions must have to account for the combined effects of forward and backward masking described in my paper (Uttal, 1969). In setting the stage for his comments, however, he fails to consider other statements made in that paper.

The joint effect of forward and backward visual masking appears to be more extended in time than that predicted from an algebraic summation of individual forward and backward masking effects. It is suggested that this apparent extension of the temporal range of masking arises from the summation of forward and backward masking effects which in themselves are insufficient to influence observed performance. Such latent masking effects are possible if it is not assumed that when stimulus identification reaches the 100% level the mask has no effect at all on the stimulus.

An investigation was conducted into the interaction of the forward and backward masking effects of unpatterned visual stimuli. It was found that detection of a test spot was easier under conditions that should have provided both forward and backward masking than under either forward masking or backward masking alone. The implications for an integration theory of masking are discussed, and the findings are contrasted with findings on the interaction of forward and backward masking by dynamic visual noise.