Cristina Rubino

Word and text processing in developmental prosopagnosia

ABSTRACT The “many-to-many” hypothesis proposes that visual object processing is supported by distributed circuits that overlap for different object categories. For faces and words the hypothesis posits that both posterior fusiform regions contribute to both face and visual word perception and predicts that unilateral lesions impairing one will affect the other. However, studies testing this hypothesis have produced mixed results. We evaluated visual word processing in subjects with developmental prosopagnosia, a condition linked to right posterior fusiform abnormalities. Ten developmental prosopagnosic subjects performed a word-length effect task and a task evaluating the recognition of word content across variations in text style, and the recognition of style across variations in word content. All subjects had normal word-length effects. One had prolonged sorting time for word recognition in handwritten stimuli. These results suggest that the deficit in developmental prosopagnosia is unlikely to affect visual word processing, contrary to predictions of the many-to-many hypothesis.

The effects of homonymous hemianopia in experimental studies of alexia

Pure alexia is characterized by an increased word-length effect in reading. However, this disorder is usually accompanied by right homonymous hemianopia, which itself can cause a mildly increased word-length effect. Some alexic studies have used hemianopic patients with modest word-length effects: it is not clear (a) whether they had pure alexia and (b) if not, whether their results could be explained by the field defect. Our goal was to determine if impairments in visual processing claimed to be related to alexia could be replicated in homonymous hemianopia alone. Twelve healthy subjects performed five experiments used in two prior studies of alexia, under both normal and simulated hemianopic conditions, using a gaze-contingent display generated by an eye-tracker. We replicated the increased word-length effect for reading time with right homonymous hemianopia, and showed a similar effect for a lexical decision task. Simulated hemianopia impaired scanning accuracy for letter or number strings, and slowed object part processing, though the effect of configuration was not greater under hemianopic viewing. Hemianopia impaired the identification of words whose letters appeared and disappeared sequentially on the screen, with better performance on a cumulative presentation in which the letters remained on the screen. The reporting of trigrams was less accurate with hemianopia, though syllabic structure did not influence the results. We conclude that some impairments that have been attributed to the processing defects underlying alexia may actually be due to right homonymous hemianopia. Our results underline the importance of considering the contribution of accompanying low-level visual impairments when studying high-level processes.

The impact of central sparing on the word-length effect in hemianopia

ABSTRACT Studies suggest that a word-length effect of up to 160 ms/letter distinguishes hemianopic dyslexia from pure alexia. However, partial preservation of central vision is common in right hemianopia, but its effects on single-word reading are unknown. Eighteen healthy subjects read single words with a gaze-contingent right hemianopia simulation that varied the degree of central sparing. Mean reading onset time declined with small degrees of central sparing, but the word-length effect did not decrease until sparing exceeded 3.15°. We next evaluated the effects of font size. Effects of central sparing were constant when expressed in number of letters, with a decline in word-length effect beginning as sparing approached 4 letters. We conclude that the effects of central sparing on mean reading onset time and the word-length effect are distinct. We provide diagnostic word-length criteria for discriminating between pure alexia and hemianopic dyslexia with various degrees of central sparing.

Valence, expression and identity effects in the affective priming paradigm

BACKGROUND In the affective-priming paradigm, a briefly-presented facial expression (prime) facilitates reaction time to identify a subsequent facial expression (target) if the prime and target are congruent in emotional valence (i.e. both positive or both negative). However, some studies have found that facilitation is more specific, only occurring when the prime and target share the same emotion (e.g. both angry). OBJECTIVE We investigated first whether reaction times to identify expressions showed valence priming effects that were modulated by whether the specific emotion was congruent between the prime and target. Second, we examined whether effects differed when the prime and target were faces of different people, a situation which minimizes low-level image effects. METHODS 22 subjects were asked to indicate if the valence of a target facial expression was positive or negative. A prime (no prime, angry, fearful, happy) was shown for 85ms, preceded and followed by masks, and then a target expression (angry, fearful, or happy) was shown for 100ms. We examined whether priming by negative emotions (angry or fearful) differed from positive emotions (happy). Next we assessed whether there was an interaction between specific expressions of the prime and of the target. RESULTS We replicated the valence effect of priming, showing that subjects were faster to respond to positive emotions with a positive prime, and negative emotions with negative primes. Similar effects were obtained when the faces of the prime and target differed in identity. Examination of effects on negative targets showed no interaction between prime and target, in either same or different-identity blocks. CONCLUSION Expression-priming effects are related to emotional valence rather than specific expressions. Along with the finding that the effects do not depend on facial identity, this indicates a high-level origin of the priming effect, rather than an effect generated by low-level image properties. Meeting abstract presented at VSS 2015.

Interactions between the perception of age and ethnicity in faces: an event-related potential study

Face perception models propose that different facial attributes are processed by anatomically distinct neural pathways that partially overlap. Whether these attributes interact functionally is an open question. Our goal was to determine if there are interactions between age and ethnicity processing and, if so, at what temporal epoch these interactions are evident. We monitored event-related potentials on electroencephalography while subjects categorized faces by age or ethnicity in two conditions: a baseline in which the other of these two properties not being categorized was held constant and an interference condition in which it also varied, as modelled after the Garner interference paradigm. We found that, when participants were categorizing faces by age, variations in ethnicity increased the amplitude of the right face-selective N170 component. When subjects were categorizing faces by ethnicity, variations in age did not alter the N170. We concluded that there is an asymmetric pattern of influence between age and ethnicity on early face-specific stages of visual processing, which has parallels with behavioural evidence of asymmetric interactions between identity and expression processing of faces.