A long-lasting dispute within typography is whether serif or sans serif fonts are the most legible. However, several parameters affect font legibility; not just whether serifs are present or not. An experiment was conducted to test whether reading performance was more affected by serifs or by the contrast of the font's letter-stroke. We created four fonts, such that their stylistic features were isolated; serifs and letter-stroke contrast. Type of serif was manipulated to be serif or sans serif, letter-stoke contrast was manipulated to be low or high (i.e., 2 by 2, factorial design), and a lexical decision task was implemented. That is, participants were exposed to a pair of words, one after the other (separate intervals), and one word was always misspelled (swapped two of its middle letters). Participants indicated in which of the two intervals the correctly spelled word was placed (i.e., two-interval, forced-choice task). As a baseline condition, three lexical identification thresholds were obtained via the QUEST algorithm and the font Helvetica. These three thresholds were averaged and used to tailor each participant's stimuli intensities, which were 9 levels of font size. The method of constant stimuli was implemented where the font size varied each trial. There was a main effect of letter-stroke contrast such that low-contrast letter-stroke elicited lower lexical identification thresholds (M = 61.12 points), relative to those elicited by high-contrast letter-stroke (M = 63.49 points), F (1, 14) = 5.36, p = .036. Neither the main effect of type of serif nor its interaction with letter-stroke contrast were significant, which indicated that the serif parameter did not significantly influence lexical identification. These results lend support for an account through which letter-stroke with low contrast, and not serifs, facilitates observers' lexical identification thresholds.

Designing legible fonts often involves balancing various trade-offs. While the added negative space that surrounds light-weight fonts enhances legibility by mediating the effects of crowding from nearby letters (Dobres, Reimer, & Chahine, 2016), it also impedes legibility at small visual angles by taking up the black space needed for letter recognition at low spatial frequencies (Beier & Oderkerk, 2019). Given the dependence of crowding on visual complexity (Bernard & Chung, 2011) and that stroke frequency has been shown to be a predictor of spatial frequency channels, here, we facilitated letter recognition at small visual angles by increasing the letter width in order to decrease visual complexity by way of the stroke frequency, defined as the number of lines crossed by a horizontal slice through a letter, divided by the width of that letter (Majaj, Pelli, Kurshan, & Palomares, 2002). To investigate the effect of width on letter recognition, we tested three variations of Helvetica (Condensed, Roman, and Expanded). We employed a short exposure single report trigram paradigm in which a string of three letters were presented left or right of the center in the fovea at 1.3° eccentricity. Participants were instructed to report the middle letter while maintaining fixation on the fixation cross. We used an adapted accelerated staircase procedure to determine individual stimulus x-heights and used backward masking to control the exposure duration. A repeated-measures ANOVA indicated a main effect of font width. Bonferroni corrected post hoc t-tests showed that recognition performance decreased significantly with width. Specifically, mean recognition for Roman was significantly higher than for Condensed, while recognition for Expanded was significantly higher than both Roman and Condensed. The findings indicate that wider letter shapes improve reading at small visual angles.

Accurate temporal expectation about an event should reflect its hazard rate, i.e., the probability it will occur in the next possible moment, given that it hasn't yet. Indeed, the speed of perceptual processing of visual stimuli has been found to change as a function of the hazard rate of those stimuli when varied between blocks of trials in visual recognition paradigms (Vangkilde, Coull, & Bundesen, 2012). In this study we investigated the dynamic adaptation of temporal expectation over time. We used two alternating hazard rates over the course of a variable cue-stimulus foreperiod, and measured participant response times in a speeded visual discrimination task. During the foreperiod, a coloured fixation cross would alternate between two colours, each denoting either a high or a low constant probability that the target stimulus would appear in the next possible moment. Foreperiods were distributed using discrete time steps of 400 ms, each with a high or a low probability that the stimulus would appear on that time step. Consecutive time steps with the same hazard rate were grouped together into epochs such that the probability that the stimulus would appear during the next time step was dependent on whether that time step was located during a high or a low probability epoch. We found significantly faster responses to stimuli presented during high probability epochs than during low probability epochs. This indicates that participants were not only able to adjust their temporal expectation, but could do so fast and flexible enough to reflect the change in hazard rate. Meeting abstract presented at VSS 2017

Rhythm has often been demonstrated to facilitate both motor performance and perceptual processing within the auditory domain. Miller, Carlson, and McAuley (2013) represents one of the few examples of facilitated report accuracy of visual stimuli presented in synchrony with an entrained rhythm. In a series of studies, we investigated the effect of temporal expectancy on visual perception, induced by an auditory rhythmic entrainment paradigm ad modum Miller et al. To obtain estimates of perceptual performance unconfounded by motor components, we employed non-speeded accuracy-based measures. In the first experiment, participants were presented with a series of seven rhythmic tones, before being shown a single visual stimulus in, or out of synch with the entraining rhythm. Exposure durations of the stimuli were systematically varied and estimates of perceptual processing speed and the threshold for conscious perception were derived using computational modelling based on Bundesen's Theory of Visual Attention (TVA; 1990). Contrary to Miller et al. our findings indicated lower processing speeds only for stimuli presented earlier than expected. This indicates a ramping up of expectancy caused by a failure to control for foreperiod effects and supports findings by Vangkilde, Petersen, and Bundesen (2014), who showed that the speed at which visual stimuli were processed increased with temporal expectation. However, further studies in which we controlled for expectancy by varying the length of the entraining period, also did not show an accuracy benefit for visual stimuli presented in synch with an entraining rhythm. Our findings and the robust non-replication are discussed in relation to the existing rhythmic expectancy literature. Meeting abstract presented at VSS 2016

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ? Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.