Myriam Chanceaux

The influence of clutter on real-world scene search: Evidence from search efficiency and eye movements

We investigated the relationship between visual clutter and visual search in real-world scenes. Specifically, we investigated whether visual clutter, indexed by feature congestion, sub-band entropy, and edge density, correlates with search performance as assessed both by traditional behavioral measures (response time and error rate) and by eye movements. Our results demonstrate that clutter is related to search performance. These results hold for both traditional search measures and for eye movements. The results suggest that clutter may serve as an image-based proxy for search set size in real-world scenes.

Serial position effects in the identification of letters, digits, symbols, and shapes in peripheral vision

Three experiments measured serial position functions for character-in-string identification in peripheral vision. In Experiment 1, random strings of five letters (e.g., P F H T M) or five symbols (e.g., λ Б Þ Ψ ¥) were briefly presented to the left or to the right of fixation, and identification accuracy was measured at each position in the string using a post-cued two-alternative forced-choice task (e.g., was there a T or a B at the 4th position). In Experiment 2 the performance to letter stimuli was compared with familiar two-dimensional shapes (e.g., square, triangle, circle), and in Experiment 3 we compared digit strings (e.g., 6 3 7 9 2) with a set of keyboard symbols (e.g., % § @ < ?). Eye-movements were monitored to ensure central fixation. The results revealed a triple interaction between the nature of the stimulus (letters/digits vs. symbols/shapes), eccentricity, and visual field. In all experiments this interaction reflected a selective left visual field advantage for letter or digit stimuli compared with symbol or shape stimuli for targets presented at the greatest eccentricity. The results are in line with the predictions of the modified receptive field hypothesis proposed by Tydgat and Grainger (2009), and the predictions of the SERIOL2 model of letter string encoding.

A Computational Cognitive Model of Information Search in Textual Materials

Document foraging for information is a crucial and increasingly prevalent activity nowadays. We designed a computational cognitive model to simulate the oculomotor scanpath of an average web user searching for specific information from textual materials. In particular, the developed model dynamically combines visual, semantic, and memory processes to predict the user’s focus of attention during information seeking from paragraphs of text. A series of psychological experiments was conducted using eye-tracking techniques in order to validate and refine the proposed model. Comparisons between model simulations and human data are reported and discussed taking into account the strengths and shortcomings of the model. The proposed model provides a unique contribution to the investigation of the cognitive processes involved during information search and bears significant implications for web page design and evaluation.

Effects of number, complexity, and familiarity of flankers on crowded letter identification

We tested identification of target letters surrounded by a varying number (2, 4, 6) of horizontally aligned flanking elements. Strings were presented left or right of a central fixation dot, and targets were always at the center of the string. Flankers could be other letters, digits, symbols, simple shapes, or false fonts, and thus varied both in terms of visual complexity and familiarity. Two-alternative forced choice (2AFC) speed and accuracy was measured for choosing the target letter versus an alternative letter that was not present in the string. Letter identification became harder as the number of flankers increased. Greater flanker complexity led to more interference in target identification, whereas more complex targets were easier to identify. Effects of flanker complexity were found to depend on visual field and position of flankers, with the strongest effects seen for leftward flankers in the left visual field. Visual complexity predicted flanker interference better than familiarity, and better than target-flanker similarity. These results provide further support for an excessive feature-integration account of the interfering effects of both adjacent and nonadjacent flanking elements in horizontally aligned strings.

Word processing speed in peripheral vision measured with a saccadic choice task

A saccadic choice task (Kirchner & Thorpe, 2006) was used to measure word processing speed in peripheral vision. To do so, word targets were accompanied by distractor stimuli, which were random strings of consonants presented in the contralateral visual field. Participants were also tested with the animal stimuli of Kirchner and Thorpes original study. The results obtained with the animal stimuli provide a straightforward replication of prior findings, with the estimated fastest saccade latencies to animal targets being 140 ms. With the word targets, the fastest reliable saccades occurred with latencies of around 200 ms. The results obtained with word targets provide a timing estimate for word processing in peripheral vision that is incompatible with sequential-attention-shift (SAS) accounts of eye movement control in reading.

A Model to Simulate Web Users' Eye Movements

One of the most important tasks on the Web is foraging information. In this study, we present a computational model which simulates human eye movements during information seeking in Web pages. Human visual scanpaths are guided by their information needs, by the visual features of the stimuli and by what they previously processed. Our model takes into account both semantic (top-down) and visual (bottom-up) information, as well as a memory model in order to predict the focus of attention. Our model operates at the block level, but also at the word level. To validate this model, both participants and model were asked to seek information in a pseudo online newspaper. We find good correspondence between simulated and empirically observed scanpaths. Knowing where the user is looking at while searching for information is crucial for evaluating the usability of Web sites and contribute to the design of Web pages.

Towards a Model of Information Seeking by Integrating Visual, Semantic and Memory Maps

This paper presents a threefold model of information seeking. A visual, a semantic and a memory map are dynamically computed in order to predict the location of the next fixation. This model is applied to a task in which the goal is to find among 40 words the one which best corresponds to a definition. Words have visual features and they are semantically organized. The model predicts scanpaths which are compared to human scanpaths on 3 high-level variables (number of fixations, average angle between saccades, rate of progression saccades). The best fit to human data is obtained when the memory map is given a strong weight and the semantic component a low weight.

Constraints on Letter-in-String Identification in Peripheral Vision: Effects of Number of Flankers and Deployment of Attention

Effects of non-adjacent flanking elements on crowding of letter stimuli were examined in experiments manipulating the number of flanking elements and the deployment of spatial attention. To this end, identification accuracy of single letters was compared with identification of letter targets surrounded by two, four, or six flanking elements placed symmetrically left and right of the target. Target stimuli were presented left or right of a central fixation, and appeared either unilaterally or with an equivalent number of characters in the contralateral visual field (bilateral presentation). Experiment 1A tested letter targets with random letter flankers, and Experiments 1B and 2 tested letter targets with Xs as flanking stimuli. The results revealed a number of flankers effect that extended beyond standard two-flanker crowding. Flanker interference was stronger with random letter flankers compared with homogeneous Xs, and performance was systematically better under unilateral presentation conditions compared with bilateral presentation. Furthermore, the difference between the zero-flanker and two-flanker conditions was significantly greater under bilateral presentation, whereas the difference between two-flankers and four-flankers did not differ across unilateral and bilateral presentation. The complete pattern of results can be captured by the independent contributions of excessive feature integration and deployment of spatial attention to letter-in-string visibility.

Deciphering CAPTCHAs: what a Turing test reveals about human cognition.

Turning Turings logic on its head, we used widespread letter-based Turing Tests found on the internet (CAPTCHAs) to shed light on human cognition. We examined the basis of the human ability to solve CAPTCHAs, where machines fail. We asked whether this is due to our use of slow-acting inferential processes that would not be available to machines, or whether fast-acting automatic orthographic processing in humans has superior robustness to shape variations. A masked priming lexical decision experiment revealed efficient processing of CAPTCHA words in conditions that rule out the use of slow inferential processing. This shows that the human superiority in solving CAPTCHAs builds on a high degree of invariance to location and continuous transforms, which is achieved during the very early stages of visual word recognition in skilled readers.

The World (of Warcraft) through the eyes of an expert

Negative correlations between pupil size and the tendency to look at salient locations were found in recent studies (e.g., Mathôt et al., 2015). It is hypothesized that this negative correlation might be explained by the mental effort put by participants in the task that leads in return to pupil dilation. Here we present an exploratory study on the effect of expertise on eye-movement behavior. Because there is no available standard tool to evaluate WoW players’ expertise, we built an off-game questionnaire testing players’ knowledge about WoW and acquired skills through completed raids, highest rated battlegrounds, Skill Points, etc. Experts (N = 4) and novices (N = 4) in the massively multiplayer online role-playing game World of Warcraft (WoW) viewed 24 designed video segments from the game that differ in regards with their content (i.e, informative locations) and visual complexity (i.e, salient locations). Consistent with previous studies, we found a negative correlation between pupil size and the tendency to look at salient locations (experts, r = − .17, p < .0001, and novices, r = − .09, p < .0001). This correlation has been interpreted in terms of mental effort: People are inherently biased to look at salient locations (sharp corners, bright lights, etc.), but are able (i.e., experts) to overcome this bias if they invest sufficient mental effort. Crucially, we observed that this correlation was stronger for expert WoW players than novice players (Z = − 3.3, p = .0011). This suggests that experts learned to improve control over eye-movement behavior by guiding their eyes towards informative, but potentially low-salient areas of the screen. These findings may contribute to our understanding of what makes an expert an expert.