Measures of angularity in digital images.

In light of the growing interest in studying the affective and aesthetic attributes of curvature, the present paper describes four digital image processing techniques that can be used to objectively discriminate between angular and curvilinear stimuli. MATLAB scripts for each of the techniques accompany the paper. Three studies are then reported that evaluate the efficacy of five metrics, derived from the four techniques, at quantifying the degree of angularity depicted in an image. Images of simple polygons (Study 1), artistic drawings of everyday objects (Study 2), and real-world objects, typefaces, and abstract patterns (Study 3) were analyzed. Logistic regression models were used to determine the relative importance of the metrics at distinguishing between angular and curvilinear items. With one exception, all of the metrics were capable of distinguishing between angular and curvilinear items at a level above chance, but some metrics were better at doing so than others, and their discriminative capacity was influenced by the characteristics of the image. The strengths and limitations of the metrics are discussed, as well as some practical recommendations.

Different measures of holistic face processing tap into distinct but partially overlapping mechanisms.

Holistic processing, which includes the integration of facial features and analysis of their relations to one another, is a hallmark of what makes faces 'special'. Various experimental paradigms purport to measure holistic processing but these have often produced inconsistent results. This has led researchers to question the nature and structure of the mechanism(s) underlying holistic processing. Using an individual differences approach, researchers have examined relations between various measures of holistic processing in an attempt to resolve these questions. In keeping with this, we examined relationships between four commonly used measures of holistic face processing in a large group of participants (N = 223): (1) The Face Inversion Effect, (2) the Part Whole Effect (PWE), (3) the Composite Face Effect, and (4) the Configural Featural Detection Task (CFDT). Several novel methodological and analytical elements were introduced, including the use of factor analysis and the inclusion of control conditions to confirm the face specificity of all of the effects measured. The four indexes of holistic processing derived from each measure loaded onto two factors, one encompassing the PWE and the CFDT, and one encompassing the CE. The 16 conditions tested across the four tasks loaded onto four factors, each factor corresponding to a different measure. These results, together with those of other studies, suggest that holistic processing is a multifaceted construct and that different measures tap into distinct but partially overlapping elements of it.

Directed forgetting of emotionally valenced faces.

An item-method directed forgetting task was used in three studies to present photographs of happy, neutral and sad faces to participants who had been induced to adopt a happy, neutral or sad mood. At test remember, forget or new judgments of old and new photographs of happy, neutral or sad faces were collected. According to the affect-as-cognitive-feedback hypothesis positively valenced stimuli serve as 'go signals' validating the use of currently accessible cognitions to process task demands whereas negatively valenced stimuli serve as 'stop signals' inhibiting or reversing the use of those cognitions. Since directed forgetting tasks entail the cognitions (among others) that some stimuli should be remembered and others should be forgotten, happy faces should facilitate task demands whereas sad faces should not. As predicted, directed forgetting effects were found for happy but not sad faces in Experiments 1 and 3, and directed forgetting effects were found neutral valenced faces in Experiment 2. Across all three studies mood state did not influence directed forgetting. Findings are discussed in terms of the effects of facial valence cues on directed forgetting and some directions for future research.

Spatial Frequency Thresholds for Detecting Latent Facial Signals of Threat.

This study examined whether latent facial signals of threat can be detected at more extreme ranges of spatial frequencies (SFs), and thus with fewer frequencies from an optimal middle band for face identification, compared with latent nonthreatening facial signals. Using an adaptive staircase procedure and a two-interval forced-choice same-different task, SF thresholds from the lower and higher ends of the SF spectrum were obtained for nonexpressive threatening and nonthreatening faces. Threatening faces were discriminated from neutral faces more quickly and accurately, and engendered more extreme SF thresholds, compared with nonthreatening faces. The results indicate that the components of latent threatening facial signals can be detected under a greater degree of impoverished visual information for face processing compared with their nonthreatening counterparts.

The Saliency of Angular Shapes in Threatening and Nonthreatening Faces.

Several lines of evidence suggest that angularity and curvilinearity are relied upon to infer the presence or absence of threat. This study examines whether angular shapes are more salient in threatening compared with nonthreatening emotionally neutral faces. The saliency of angular shapes was measured by the amount of local maxima in S(θ), a function that characterizes how the Fourier magnitude spectrum varies along specific orientations. The validity of this metric was tested and supported with images of threatening and nonthreatening real-world objects and abstract patterns that have predominantly angular or curvilinear features (Experiment 1). This metric was then applied to computer-generated faces that maximally correlate with threat (Experiment 2a) and to real faces that have been rated according to threat (Experiment 3). For computer-generated faces, angular shapes became increasingly salient as the threat level of the faces increased. For real faces, the saliency of angular shapes was not predictive of threat ratings after controlling for other well-established threat cues, however, other facial features related to angularity (e.g., brow steepness) and curvilinearity (e.g., round eyes) were significant predictors. The results offer preliminary support for angularity as a threat cue for emotionally neutral faces.

Effects of Photo-Depicted Pupil Diameter on Judgments of Others' Attentiveness and on Facial Recognition Memory.

Occasionally, individuals perceive that someone is no longer paying attention to the discussion at hand even when there are no overt cues of inattentiveness. As a preliminary study of this phenomenon, we examined whether pupil diameter might be implicitly used to infer others' attentiveness. Forty participants (27 women, 13 men, M age = 19.7 year, SD = 2.8) were presented with images of male faces with either large or small pupils, and, in the context of a personnel selection scenario, participants then judged the attentiveness of the person in the image. Images of faces with large pupils were judged as more attentive, compared with images of faces with small pupils. Face recognition memory performance was not affected by depicted pupil size. Our results are consistent with the proposal that pupillary fluctuations can be an index of perceived attention, and they provide preliminary evidence that pupil dilation may be implicitly relied upon to infer attentional states.

Configural and featural discriminations use the same spatial frequencies: a model observer versus human observer analysis.

Previous work has shown mixed results regarding the role of different spatial frequency (SF) ranges in featural and configural processing of faces. Some studies suggest no special role of any given band for either type of processing, while others suggest that low SFs principally support configural analysis. Here we attempt to put this issue on a more rigorous footing by comparing human performance when making featural and configural discriminations with that of a model observer algorithm carrying out the same task. The model uses a simple algorithm that calculates the dot product of a stimulus image with each available potential match image to find the maximally likely match. It thus provides a principled way of analyzing available image information. We find human accuracy peaks at around 10 cycles per face (cpf) regardless of whether featural or configural manipulations are being detected. We also find accuracy peaks in the same part of the spectrum regardless of which feature is manipulated (ie eyes, nose, or mouth). Conversely, model observer performance, measured in terms of white noise tolerance, peaks at approximately 5 cpf, and this value again remains roughly constant regardless of the type of manipulation and feature manipulated. The ratio of the model's noise tolerance to a derived equivalent noise tolerance value for humans peaks at around 10 cpf, similar to the accuracy data. These results provide evidence that the human performance maxima at 10 cpf are not due simply to the physical characteristics of face stimuli, but rather arise due to an interaction between the available information in face images and human perceptual processing.

Depth of treatment sensitive noise resistant dynamic artificial neural networks model of recall in people with prosopagnosia.

The Fusiform Face Area (FFA) is the brain region considered to be responsible for face recognition. Prosopagnosia is a brain disorder causing the inability to a recognise faces that is said to mainly affect the FFA. We put forward a model that simulates the capacity to retrieve label associated with faces and objects depending on the depth of treatment of the information. Akin to prosopagnosia, various localised "lesions" were inserted into the network in order to evaluate the degradation of performance. The network is first composed of a Feature Extracting Bidirectional Associative Memory (FEBAM-SOM) to represent the topological maps allowing the categorisation of all faces. The second component of the network is a Bidirectional Heteroassociative Memory (BHM) that links those representations to their semantic label. For the latter, specific semantic labels were used as well as more general ones. The inputs were images representing faces and various objects. Just like in the visual perceptual system, the images were pre-processed using a low-pass filter. Results showed that the network is able to associate the extracted map with the correct label information. The network is able to generalise and is robust to noise. Moreover, results showed that the recall performance of names associated with faces decrease with the size of lesion without affecting the performance of the objects. Finally, results obtained with the network are also consistent with human ones in that higher level, more general labels are more robust to lesion compared to low level, specific labels.

The effects of distinctiveness on memory and metamemory for face-name associations.

We examined the influence of face and name distinctiveness on memory and metamemory for face-name associations. Four types of monitoring judgements were solicited during encoding and retrieval of face-name pairs that contained distinct or typical faces (Experiment 1) or names (Experiment 2). The beneficial effects of distinctiveness on associative memory were symmetrical between faces and names, such that relative to their typical counterparts, distinct faces enhanced memory for names, and distinct names enhanced memory for faces. These effects were also apparent in metamemory. Estimates of prospective and retrospective memory performance were greater for face-name associations that contained a distinct face or name compared with a typical face or name, regardless of whether the distinct item was a cue or target. Moreover, the predictive validity of prospective monitoring improved with name distinctiveness, whereas the predictive validity of retrospective monitoring improved with facial distinctiveness. Our results indicate that distinctiveness affects not only the strength of the association between a face and a name, but also the ability to monitor that association.

Metamemory for faces, names, and common nouns.

This study examined the metacognitive aspects of face-name learning with the goal of providing a comprehensive profile of monitoring performance during this task. Four types of monitoring judgments were solicited during encoding and retrieval of novel face-name associations. Across all of the monitoring judgments, relative accuracy was significantly above chance for face and name targets. Furthermore, metamemory performance was similar between both target conditions, even though names were more difficult to recognize than faces. As a preliminary test of the stability of monitoring accuracy across different categories of stimuli, we also compared metamemory performance between face-name pairs and noun-noun pairs. Prospective monitoring accuracy was similar across the categories of stimuli, but retrospective monitoring accuracy was superior for noun targets compared with face or name targets. Altogether, our results indicate that participants can monitor their memory for face-name associations at a level above chance, and retrospective monitoring is more accurate with nouns compared with faces and names.

Spatial-frequency thresholds for configural and featural discriminations in upright and inverted faces.

Face recognition is thought to rely more on the relative positions of face features (configural information) than on the appearance of the individual face parts (featural information). It also seems to rely on a specific band of spatial frequencies (SFs). In this study, we measured the SFs needed for processing configural and featural information using the method of constant stimuli in combination with a simultaneous-matching paradigm. Stimuli were two-octave-wide bandpass-filtered upright and inverted faces that contained either featural or configural modifications. SF thresholds for featural and configural processing were calculated by interpolating between discrimination accuracy scores. Low-pass and high-pass thresholds for featural and configural processing in upright faces were approximately equal, whereas for inverted faces, the thresholds were closer to the middle of the spectrum for configural processing relative to featural processing. Thus, a broader band of SFs, one that overlapped more with the middle of the frequency spectrum, was needed for configural processing than for featural processing in inverted faces. Our findings emphasise the importance of a narrow mid-range band of frequencies for both configural and featural encoding in upright faces and suggest that configural information is extracted less effectively in inverted faces.

Effects of familiarity on spatial frequency thresholds for face matching.

We examined whether familiarity with a face influences the spatial frequencies (SFs) required for face matching. Using the psychophysical method of constant stimuli and a 3AFC simultaneous matching paradigm, we obtained SF thresholds for familiar- and unfamiliar-face matching from fourteen observers, of which four were personally familiar with a subset of the faces while the remainder served as controls. SF thresholds from the lower extreme of the spectrum were approximately one octave lower for familiar than for unfamiliar faces, while SF thresholds from the upper extreme of the spectrum were approximately a third of an octave higher. These results highlight a quantitative difference between processing familiar and unfamiliar faces.