Longer Presentation Durations Research Articles

Longer presentation duration helps to individuate faces in an RSVP stream

Longer presentation duration helps to individuate faces in an RSVP stream

Mechanisms of human dynamic object recognition revealed by sequential deep neural networks.

Humans can quickly recognize objects in a dynamically changing world. This ability is showcased by the fact that observers succeed at recognizing objects in rapidly changing image sequences, at up to 13 ms/image. To date, the mechanisms that govern dynamic object recognition remain poorly understood. Here, we developed deep learning models for dynamic recognition and compared different computational mechanisms, contrasting feedforward and recurrent, single-image and sequential processing as well as different forms of adaptation. We found that only models that integrate images sequentially via lateral recurrence mirrored human performance (N = 36) and were predictive of trial-by-trial responses across image durations (13-80 ms/image). Importantly, models with sequential lateral-recurrent integration also captured how human performance changes as a function of image presentation durations, with models processing images for a few time steps capturing human object recognition at shorter presentation durations and models processing images for more time steps capturing human object recognition at longer presentation durations. Furthermore, augmenting such a recurrent model with adaptation markedly improved dynamic recognition performance and accelerated its representational dynamics, thereby predicting human trial-by-trial responses using fewer processing resources. Together, these findings provide new insights into the mechanisms rendering object recognition so fast and effective in a dynamic visual world.

Enumerating the forest before the trees: The time courses of estimation-based and individuation-based numerical processing

Ensemble perception refers to the ability to report attributes of a group of objects, rather than focusing on only one or a few individuals. An everyday example of ensemble perception is the ability to estimate the numerosity of a large number of items. The time course of ensemble processing, including that of numerical estimation, remains a matter of debate, with some studies arguing for rapid, “preattentive” processing and other studies suggesting that ensemble perception improves with longer presentation durations. We used a forward-simultaneous masking procedure that effectively controls stimulus durations to directly measure the temporal dynamics of ensemble estimation and compared it with more precise enumeration of individual objects. Our main finding was that object individuation within the subitizing range (one to four items) took about 100–150 ms to reach its typical capacity limits, whereas estimation (six or more items) showed a temporal resolution of 50 ms or less. Estimation accuracy did not improve over time. Instead, there was an increasing tendency, with longer effective durations, to underestimate the number of targets for larger set sizes (11–35 items). Overall, the time course of enumeration for one or a few single items was dramatically different from that of estimating numerosity of six or more items. These results are consistent with the idea that the temporal resolution of ensemble processing may be as rapid as, or even faster than, individuation of individual items, and support a basic distinction between the mechanisms underlying exact enumeration of small sets (one to four items) from estimation.

Neural reinstatement and the amount of information recollected

Recent functional neuroimaging studies have attempted to understand the cognitive and neural bases of episodic memory retrieval, as well as the extent to which different retrieval judgments reflect qualitative as opposed to continuous changes in neural signals. The present study addressed this issue by investigating the reinstatement of episodic content according to the amount of information available at retrieval. Subjects encoded a series of words in the context of three distinct tasks, while a manipulation of presentation duration (4 or 8s) was also employed. A later recognition memory test was used to segregate trials according to whether or not they were accompanied by the recollection of details from encoding. Functional MRI data acquired during both the encoding and retrieval phases were used in conjunction with multi-voxel pattern-analysis (MVPA) to provide a measure of the degree to which encoding-related patterns of brain activity were later reactivated (reinstated) at the time of retrieval. Critically, the magnitude of reinstatement differed with respect to the encoding manipulation, such that reinstatement was stronger for items associated with the longer presentation duration. Together with duration-related differences in retrieval activity in left posterior parietal cortex, the results provide neural evidence for the reinstatement of different amounts of episodic information, consistent with the idea that recollection is based on a continuous neural signal.

The speed and accuracy of material recognition in natural images

We studied the time course of material categorization in natural images relative to superordinate and basic-level object categorization, using a backward-masking paradigm. We manipulated several low-level features of the images-including luminance, contrast, and color-to assess their potential contributions. The results showed that the speed of material categorization was roughly comparable to the speed of basic-level object categorization, but slower than that of superordinate object categorization. The performance seemed to be crucially mediated by low-level factors, with color leading to a solid increase in performance for material categorization. At longer presentation durations, material categorization was less accurate than both types of object categorization. Taken together, our results show that material categorization can be as fast as basic-level object categorization, but is less accurate.

Using an Oculomotor Signature as an Indicator of Aesthetic Preference

Eye movements are strongly influenced by the task given to an observer. The immediacy of such eye movements, which are difficult to control consciously, offers the potential to explore highly variable subjective evaluations, such as aesthetic preference, with reliable objective measures. We presented a variety of images in sets of 2, 4, or 8 items for different durations and analyzed oculomotor statistics such as cumulative fixation duration, refixations, and the sequence of fixations while participants searched for their preferred image, after which participants indicated their preference using a button press. The total amount of time spent looking at any image correlates with selection preference and does so increasingly well with longer presentation duration. For short presentations, the first and last fixations correlate better with image preference. All response measures become increasingly variable as the number and complexity of presented images are increased. A weighted combination of these measures can significantly improve the correlation with preference, suggesting a “signature” which could be used as a reliable indicator for task-free subjective evaluation of stimuli in visual psychophysics. Its role as an improved fitness function in visually driven evolutionary algorithms is discussed.

Faces versus patterns: Exploring aesthetic reactions using facial EMG.

We used facial EMG to examine reactions to the attractiveness of natural (faces) and artificial (abstract patterns) stimuli under long and short presentation durations. Attractive stimuli produced strong activations of the M. zygomaticus major muscle, indicating positive affective reactions; and unattractive stimuli produced strong activations of the M. corrugator supercili muscle, indicating negative affective reactions. Fluency effects, indicated by stronger activations of the M. zygomaticus major under the longer presentation duration were, however, only found for the abstract patterns. Moreover, the abstract patterns also were associated with more consistent activations over time than the faces, suggesting differences in the processes underlying the evaluation of faces and patterns. We discuss these results in terms of differences in appraisal processes between the two classes of stimuli—the greater biological, social, and sociosexual significance of faces trigger more complex appraisals than the abstract patterns.

Keep your eyes on the ball: smooth pursuit eye movements enhance prediction of visual motion

Success of motor behavior often depends on the ability to predict the path of moving objects. Here we asked whether tracking a visual object with smooth pursuit eye movements helps to predict its motion direction. We developed a paradigm, "eye soccer," in which observers had to either track or fixate a visual target (ball) and judge whether it would have hit or missed a stationary vertical line segment (goal). Ball and goal were presented briefly for 100-500 ms and disappeared from the screen together before the perceptual judgment was prompted. In pursuit conditions, the ball moved towards the goal; in fixation conditions, the goal moved towards the stationary ball, resulting in similar retinal stimulation during pursuit and fixation. We also tested the condition in which the goal was fixated and the ball moved. Motion direction prediction was significantly better in pursuit than in fixation trials, regardless of whether ball or goal served as fixation target. In both fixation and pursuit trials, prediction performance was better when eye movements were accurate. Performance also increased with shorter ball-goal distance and longer presentation duration. A longer trajectory did not affect performance. During pursuit, an efference copy signal might provide additional motion information, leading to the advantage in motion prediction.

Metacognitive control over the distribution of practice: When is spacing preferred?

The authors clarify the source of a conflict between previous findings related to metacognitive control over the distribution of practice. In a study by L. Son (2004), learners were initially presented pairs of Graduate Record Examination (GRE) vocabulary words and their common synonyms for 1 s, after which they chose to study the pair again immediately (massed practice), later (spaced practice), or not at all (done). Learners chose spaced practice less as pair difficulty increased. A. S. Benjamin and R. D. Bird (2006), using different materials and procedures and a longer presentation duration (5 s), concluded just the opposite. The authors adopted Son's materials and procedures and replicated her findings with a 1-s stimulus duration. However, the declining choice of spacing as item difficulty increased largely reflected learners' failure to fully perceive items with brief presentations. With longer presentations, ensuring full perception, the choice of spaced practice increased with greater pair difficulty, in agreement with Benjamin and Bird. Theoretical implications are discussed in the context of discrepancy-reduction and proximal-learning perspectives.

The influence of color on the perception of scene gist.

In 3 experiments the authors used a new contextual bias paradigm to explore how quickly information is extracted from a scene to activate gist, whether color contributes to this activation, and how color contributes, if it does. Participants were shown a brief presentation of a scene followed by the name of a target object. The target object could be consistent or inconsistent with scene gist but was never actually present in the scene. Scene gist activation was operationalized as the degree to which participants respond "yes" to consistent versus inconsistent objects, reflecting a response bias produced by scene gist. Experiment 1 demonstrated that scene gist is activated after a 42-ms exposure and that the strength of the activation increases with longer presentation durations. Experiments 2 and 3 explored the contribution of color to the activation of scene gist. The results revealed that color has an influence across a wide variety of scenes and is directly associated with scene gist.

Vertical object motion during horizontal ocular pursuit: compensation for eye movements increases with presentation duration

Smooth pursuit eye movements change the retinal image motion of objects in the visual field. To enable an observer to perceive the motion of these objects veridically, the visual system has to compensate for the effects of the eye movements. The occurrence of the Filehne-illusion (illusory motion of a stationary object during smooth pursuit) shows that this compensation is not always perfect. The amplitude of the illusion appears to decrease with increasing presentation durations of the stationary object. In this study we investigated whether presentation duration has the same effect when an observer views a vertically moving object during horizontal pursuit. In this case, the pursuit eye movements cause the perceived motion path to be oblique instead of vertical; this error in perceived motion direction should decrease with higher presentation durations. In Experiment 1, we found that the error in perceived motion direction indeed decreased with increasing presentation duration, especially for higher pursuit velocities. The results of Experiment 2 showed that the error in perceived motion direction did not depend on the moment during pursuit at which the stimulus was presented, suggesting that the degree of compensation for eye movements is constant throughout pursuit. The results suggest that longer presentation durations cause the eye movement signal that is used by the visual system to increase more than the retinal signal.

The effects of presentation time on the localization accuracy of 3D sound through headphones

The present study deals with the relationship between sound presentation time and localization accuracy of 3D sound presented through headphones. Huron Lake CP4 system was used for sound presentation and Flock of birds tracking system was used to register movement of the head. There were 31 volunteers (15 females and 16 males) in this experiment. Six different types of sounds were presented with the duration of 0.5, 2, 4, and 6 s. They were dog bark (0–500 Hz), coin drop with three different bandwidths (0–1650 Hz, 1650–6500 Hz, and 0–8500 Hz), alarm (0–5500 Hz) and male speech voice (0–7000 Hz). The results showed that the accuracy of localizing the sound is significantly related to the duration of the sound presentation. The perceived azimuth difference for 0.5 s duration (45°±15° in average) is significantly higher compared with the longer presentation duration. The perceived azimuth difference for 2 s duration is also significantly higher than 4 and 6 s duration (21°±9° in average). There is not any significant difference between 4 and 6 s duration. There is not any significant difference between male and female subjects, but individual difference is significant.

The Rise and Fall of False Recall: The Impact of Presentation Duration

The effect of presentation duration on false recall induced by presentation of semantically associated and phonologically associated word lists was examined. When 16-word lists comprised of semantically related words (e.g., bed, yawn) were presented at short durations (i.e., 20 or 250 ms/word), false recall of related but nonpresented words (e.g., sleep) increased with increasing duration. However, at longer presentation durations (i.e., 1000, 3000, or 5000 ms/word), false recall declined with increasing presentation duration. This pattern resolves discrepancies among previous experiments investigating the effects of presentation duration on associatively induced false recall. Further, these data constrain theoretical accounts of false recall in that single-process models cannot readily account for these effects. We propose a dual-process model that appears to account for these findings and much of the extant literature. Phonologically related lists (e.g., sweep, sleet) exhibited a very different pattern of results at the short presentation durations; specifically, false recall was exceedingly high at the fastest duration and declined as duration lengthened. Similarities and differences between the mechanisms underlying semantically and phonologically induced false recall are considered.

The Effects of Presentation Rate and Encoding Task on Age-Related Memory Deficits

Young and old adults studied several lists of words at each of three presentation durations (1.5, 3, and 6 s/word). Under standard, intentional learning instructions the magnitude of the age-related recall deficit increased with longer presentation durations. When a semantic orienting question was used to guide the encoding of the items on each list, however, an equivalent benefit of longer study times was observed for the two age groups. In delayed recognition testing, the age groups also showed equivalent benefits from longer study times. These results were interpreted within the context of a processing deficit framework of age differences in memory, in which age differences are due to differences in the spontaneous use of organizational strategies. The results are not well described by a cognitive slowing account of age differences in memory.