Article

Effects of image blur on visual perception and affective response

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Obscure visual environments impair visual perception and result in the worse performances in object detection, identification, and recognition. Blurred images might also induce negative affective responses to the visual environment such as fear or uncanniness. However, the effect of image blur on the affective responses is still open. The present study investigated the effects of image blur on the object detection and the feeling of uncanniness. In a psychological experiment, we presented pictures of natural images depicting a person, an animal, or an object and asked participants to indicate whether the picture depicted a face or an animal and to rate how strongly they felt the picture was uncanny. The images were blurred by the lowpass filter of various cutoff frequencies. We found that the more blurry images, namely images filtered by the lower cutoff frequencies, were more difficult in face/animal detection as well as they were rated as more uncanny. However, the effective cutoff frequencies in the modulations on the uncanniness rating and face/animal detectability did not overlap; the uncanniness and detectability did not show clear correlation. These results clearly demonstrated that the image blur elicits the negative affective responses. Furthermore, although the image blur influences both visual perception and affective response in a similar fashion, the feelings of uncanniness and face/animal detectability were not related directly from each other.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Gaussian blurring can be implemented using graphics or video-editing software, and applying this manipulation will reduce detail in an image. For instance, applying Gaussian blur to an image has been shown to negatively affect object perception speed and naming accuracy (Laws & Hunter, 2006;Takahshi & Watanabe, 2015). In Experiment 2, we evaluated the effect of reducing stimulus clarity in five steps using Gaussian blur for human stimuli only. ...
Article
Full-text available
Observing actions evokes an automatic imitative response that activates mechanisms required to execute these actions. Automatic imitation is measured using the Stimulus Response Compatibility (SRC) task, which presents participants with compatible and incompatible prompt-distractor pairs. Automatic imitation, or the compatibility effect , is the difference in response times (RTs) between incompatible and compatible trials. Past results suggest that an action’s animacy affects automatic imitation: human-produced actions evoke larger effects than computer-generated actions. However, it appears that animacy effects occur mostly when non-human stimuli are less complex or less clear. Theoretical accounts make conflicting predictions regarding both stimulus manipulations. We conducted two SRC experiments that presented participants with an animacy manipulation (human and computer-generated stimuli, Experiment 1) and a clarity manipulation (stimuli with varying visual clarity using Gaussian blurring, Experiments 1 and 2) to tease apart effect of these manipulations. Participants in Experiment 1 responded slower for incompatible than for compatible trials, showing a compatibility effect. Experiment 1 found a null effect of animacy, but stimuli with lower visual clarity evoked smaller compatibility effects. Experiment 2 modulated clarity in five steps and reports decreasing compatibility effects for stimuli with lower clarity. Clarity, but not animacy, therefore affected automatic imitation, and theoretical implications and future directions are considered.
... Examples of how scene statistics are influential include evaluating the connection between the scene and performance in specific visual tasks, predicting neural responses of a human operator, specifying how these scene features are best represented (e.g., Berman, 2018), understanding how they relate to perception (Brady et al., 2017), and understanding, as well as incorporating into design, the time course of processing associated with these different features (Mares et al., 2018). Finally, there is considerable evidence to suggest that altering these low level sensory-perceptual features can impact higher order cognition in various ways (e.g., perceptual discomfort; Habtegiorgis et al., 2019, affective response; Takahshi & Watanabe, 2015), and, of course, the most well-known examples are visual illusions (e.g., Howe & Purves 2002. ...
Chapter
The U.S. Army is currently working to integrate artificial intelligence, or AI-enabled systems, into military working teams in the form of both embodied (i.e., robotic) and embedded (i.e., computer or software) intelligent agents with the express purpose of improving performance during all phases of the mission. However, this is largely uncharted territory, making it unclear how to do this integration effectively for human-AI teams. This chapter provides an overview of the Combat Capabilities Development Command (DEVCOM) Army Research Laboratory’s effort to address the human as a critical gap with associated implications on effective teaming. This chapter articulates four major research thrusts critical to integrating AI-enabled systems into military operations, giving examples within these broader thrusts that are currently addressing specific research gaps. The four major research thrusts include: (1) Enabling Soldiers to predict AI; (2) Quantifying Soldier understanding for AI; (3) Soldier-guided AI adaptation; and (4) Characterizing Soldier-AI performance. These research thrusts are the organizing basis for explaining a path toward integration and effective human-autonomy teaming at the tactical edge.
Article
Full-text available
In the present study we examined whether categorization difficulty regarding a food is related to its likability. For this purpose, we produced stimulus images by morphing photographs of a tomato and a strawberry. Subjects categorized these images as either a tomato or a strawberry and in separate sessions evaluated the food's eatability or the subject's willingness to eat (Experiments 1 and 2) and the likeliness of existence of each food (Experiment 2). The lowest score for ca- tegorization confidence coincided with the lowest scores for eatability, willingness to eat, and likeliness of existence. In Experiment 3, we found that food neophobia, a trait of ingestion avoidance of novel foods, modulated food likability but not categorization confidence. These findings suggest that a high categorization difficulty generally co-occurs with a decrease in food likability and that food neophobia modulates likability. This avoidance of difficult-to-categorize foods seems ecologically valid because before eating we have little information regarding whether a food is potentially harmful.
Article
Full-text available
Emotional responses are regulated by basic motivational systems (appetitive and defensive), which allow for adaptive behavior when opportunities and threats are detected. It has been suggested that specific ranges of spatial frequencies of the visual input may contain information that is diagnostic for discriminating emotionally relevant from less relevant contents, and that specialized neural modules may analyze these spatial frequencies to allow efficient detection and response to life-threatening or life-sustaining stimuli. However, there is no evidence supporting this possibility regarding natural scenes, which are highly varied in terms of pictorial composition. The present study examines how low and high spatial frequency filtering affects the understanding of natural scene contents and modulates the amplitude of the late positive potential (LPP), a well-known component of the event-related potentials that reflects motivational significance. The content of an initially degraded (low- or high-passed) picture was progressively revealed in a sequence of steps by adding high or low spatial frequencies. At each step, human participants reported whether they identified the gist of the image. The results showed that the affective modulation of the LPP varied with picture identification similarly for low-passed and high-passed pictures. The engagement of corticolimbic appetitive and defensive systems, reflected in the LPP affective modulation, varied with picture identification, and did not critically or preferentially depend on either low or high spatial frequencies.
Article
Full-text available
45 undergraduates were divided into 3 equal groups differing in expectancy for receiving an unavoidable noxious stimulus on the count of 10 in a 15-number countup. One group was told that the likelihood of receiving a shock was 5%, another 50%, and another 95%. 2 trials were presented. In the 1st, no S received a shock; in the 2nd, all Ss received a shock. For both heart rate and skin conductance, the 5% expectancy group exhibited the greatest, and the 95% group the least, arousal during the anticipatory period. On both measures, the 5% expectancy group exhibited the greatest impact effect. It is concluded that within limits, low expectancy about the occurrence of an unavoidable threatening event increases anticipatory anxiety as well as the reaction to impact. (19 ref.)
Article
Full-text available
Two experiments examined the respective role of the cerebral hemispheres in face perception and the nature of their contribution depending on task demands and on the spatial-frequency composition of the stimuli. Sixteen faces of members of the subjects' department were presented as stimuli, with men and women, and professors and nonprofessors being equally represented. In Experiment 1, high-resolution black-and-white photographs of faces were used in three reaction-time tasks: verbal identification, manual membership categorization, and manual male/female categorization, in a within-subject design. Identification and membership categorization were significantly better performed in right-visual-field presentations, whereas the male/female categorization yielded a nonsignificant left-visual-field superiority. In Experiment 2, two versions of the same faces were used: digitized low-pass (0 to 2 cycles/degree of visual angle) and digitized broad-pass (0 to 32 cycles/degree) faces. Broad-pass faces produced the same laterality pattern as in Experiment 1, while low-pass faces were better processed in left-visual-field presentations for all three tasks. The results suggest that the two hemispheres play a role in face perception, and their contribution may vary as a function of the task demands and of the spatial-frequency components of the incoming information.
Chapter
This chapter introduces predictability and human stress and the notion of predictability. Two nonoverlapping classes of predictability are defined in the chapter: knowing the conditions under which the event will occur (contingency predictability) and knowing what the event will be like (what-kind-of-event predictability). The interaction of the two classes of predictability and their controllability is discussed in the chapter. The chapter describes the five major existing theories of predictability and human stress—namely, information seeking, preparatory response, preparatory set, uncontrollability, and safety signal. To specify the conditions under which predictability has stress-reducing effects and when it does not, four new hypotheses (the blunting hypothesis, the analytic hypothesis, the script hypothesis, and the arousal hypothesis) that attempt to organize the sweep of data are considered in the chapter. Humans generally prefer predictable over unpredictable aversive events. There is one exception, however—that is, when individuals can distract themselves from danger-related information, they prefer unpredictable aversive events.
Article
Human observers often experience strongly negative impressions of human-like objects falling within a particular range of visual similarity to real humans (the “uncanny valley” phenomenon). We hypothesized that negative impressions in the uncanny valley phenomenon are related to a difficulty in object categorization. We produced stimulus images by morphing two of each of real, stuffed, and cartoon human face images (Experiment 1). Observers were asked to categorize each of these images as either category and evaluated the likability of the image. The results revealed that the longest latency, the highest ambiguity in categorization, and the lowest likability score co-occurred at consistent morphing percentages. Similar results were obtained even when we employed stimulus images that were created by morphing two of each of real, stuffed, and cartoon dog images (Experiment 2). However, the effect of categorization difficulty on evaluation was weak when two real human faces were morphed (Experiment 3). These results suggest that the difficulty in categorizing an object as either of two dissimilar categories is linked to negative evaluation regardless of whether the object is human-related or not.
Article
The VideoToolbox is a free collection of two hundred C subroutines for Macintosh computers that calibrates and controls the computer-display interface to create accurately specified visual stimuli. High-level platform-independent languages like MATLAB are best for creating the numbers that describe the desired images. Low-level, computer-specific VideoToolbox routines control the hardware that transforms those numbers into a movie. Transcending the particular computer and language, we discuss the nature of the computer-display interface, and how to calibrate and control it.
Article
It is well known that object recognition requires spatial frequencies exceeding some critical cutoff value. People with central scotomas who rely on peripheral vision have substantial difficulty with reading and face recognition. Deficiencies of pattern recognition in peripheral vision, might result in higher cutoff requirements, and may contribute to the functional problems of people with central-field loss. Here we asked about differences in spatial-cutoff requirements in central and peripheral vision for letter and face recognition. The stimuli were the 26 letters of the English alphabet and 26 celebrity faces. Each image was blurred using a low-pass filter in the spatial frequency domain. Critical cutoffs (defined as the minimum low-pass filter cutoff yielding 80% accuracy) were obtained by measuring recognition accuracy as a function of cutoff frequency (in cycles per object). Our data showed that critical cutoffs increased from central to peripheral vision by 20% for letter recognition and by 50% for face recognition. We asked whether these differences could be accounted for by central/peripheral differences in the contrast sensitivity function (CSF). We addressed this question by implementing an ideal-observer model which incorporates empirical CSF measurements and tested the model on letter and face recognition. The success of the model indicates that central/peripheral differences in the cutoff requirements for letter and face recognition can be accounted for by the information content of the stimulus limited by the shape of the human CSF, combined with a source of internal noise and followed by an optimal decision rule.
Article
Recently, Tjan, Braje, Legge and Kersten [(1995) Vision Research, 35, 3053-3069] found that human efficiency for object recognition was less than 10%, indicating that humans fail to use much of the information available to an ideal observer. We examine two explanations for these low efficiencies: (1) humans are inefficient in using high spatial-frequency information; and (2) humans are inefficient in detecting image samples. We tested the first possibility by measuring human efficiency for recognizing low-pass filtered objects, rendered as line drawings and silhouettes, in luminance noise. Efficiency did not improve when high frequencies were removed, and the first explanation was rejected. We tested the second explanation by comparing efficiencies for object detection and recognition. Recognition efficiency was higher than detection efficiency for silhouettes but not line drawings, showing that detection efficiency does not place a ceiling on recognition efficiency. The results indicate that human vision is designed to extract image features, such as contours, that enhance recognition. A computer simulation suggests that this can occur if the observer views the world through a band-pass spatial-frequency channel.
Article
The color-emotion associations of undergraduate students were analyzed. Twenty men and 20 women were asked to complete a self-administered questionnaire in which they listed their favorite color, the major color they were wearing, their emotional responses to colors, and the reasons for their choices. Responses showed that bright colors elicited mainly positive emotional associations, and dark colors elicited mainly negative emotional associations. Women responded more positively than men to bright colors, and they also responded more negatively to dark colors. Comparisons are made between the color-emotion associations of children and those of adults. The reasons for the color-emotion associations are discussed.
Article
The Psychophysics Toolbox is a software package that supports visual psychophysics. Its routines provide an interface between a high-level interpreted language (MATLAB on the Macintosh) and the video display hardware. A set of example programs is included with the Toolbox distribution.
Article
Observers in recognition experiments invariably view objects against a blank background, whereas observers of real scenes sometimes view objects against dense clutter. In this study, we examined whether an object's background affects the information used for recognition. Our stimuli consisted of color photographs of everyday objects. The photographs were organized either as a sparse array, as is typical of a visual search experiment, or as high density clutter, such as might be found in a toy chest, a handbag, or a kitchen drawer. The observer's task was to locate an animal, vehicle, or food target in the stimulus. We varied the information in the stimuli by convolving them with a low-pass filter (blur) or a high-pass filter (edge) or converting them to grayscale. In two experiments, we found that the blur and edge manipulations produced a modest decrement in performance with the sparse arrangement but a severe decrement in performance with the clutter arrangement. These results indicate that the information used for recognition depends on the object's background. Thus, models of recognition that have been developed for isolated objects may not generalize to objects in dense clutter.