Discrimination of computer-synthesized speech

The effects of learning on the discrimination of computer-synthesized speech was assessed by presenting 100 computer-produced monosyllabic words to 2 groups of 15 adult subjects. One group's errors were corrected while the other group's errors were uncorrected. A comparison of errors on the first 50 vs the second 50 presentations showed significant improvement for corrected group only. It was concluded that the discrimination of computer-synthesized monosyllables can be improved with correction of errors but is still about 14% poorer than the discrimination of human speech.     

Lateral asymmetry in the perceived sequence of speech and nonspeech stimuli

The task was to estimate the position where a click had been superimposed in a spoken sentence. Experiment 1 confirmed Fodor and Bever’s observation of an ear-asymmetry effect: the click is located earlier when it is presented to the left ear and the sentence to the right ear than with the opposite arrangement. In Experiment 2, combinations of monaural and binaural presentations were considered. They made it possible to eliminate interpretations which link the laterality effect to the fact of reaching or not reaching a particular ear and showed that the relevant factor is the relative position of the stimuli in acoustic space. Experiments 3 and 4 explored the relation between spatial separation and perceived sequence in greater detail. The relation involves a plateau: when the click comes to the left of the speech, it is preposed to a degree which depends on the amount of spatial separation; but, when it comes to the right of the speech, separation is irrelevant and the mean error is of the same order of magnitude as in a control condition without separation.     

Effects of verb complexity on speech errors

Two experiments examined whether semantically related verbs that contrast with respect to the absence versus the presence of an additional semantic feature differentially compete for selection during the encoding of a sentence for production. In both experiments, a speech error induction task was used to elicit contextual (misordering) errors involving semantically related verbs that contrasted only in their semantic complexity or in both their semantic and morphophonological complexity. The prediction was that an asymmetry in contextual errors would be observed in which the more complex verbs would replace the simpler verbs more often than the reverse. This prediction was confirmed in both experiments, with more perseverations and anticipations involving the semantically more complex verb of antonym pairs in Experiment 1, and more perseverations and anticipations involving the semantically more-specified verb of a heavy-light pair in Experiment 2. The implications of the results for spreading-activation theories of language production are discussed.     

Effects of syntactic expectations on speech segmentation

Although the effect of acoustic cues on speech segmentation has been extensively investigated, the role of higher order information (e.g., syntax) has received less attention. Here, the authors examined whether syntactic expectations based on subject-verb agreement have an effect on segmentation and whether they do so despite conflicting acoustic cues. Although participants detected target words faster in phrases containing adequate acoustic cues ("spins" in take spins and "pins" in takes pins), this acoustic effect was suppressed when the phrases were appended to a plural context (those women take spins/*takes pins [with the asterisk indicating a syntactically unacceptable parse]). The syntactically congruent target ("spins") was detected faster regardless of the acoustics. However, a singular context (that woman *take spins/takes pins) had no effect on segmentation, and the results resembled those of the neutral phrases. Subsequent experiments showed that the discrepancy was due to the relative time course of syntactic expectations and acoustics cues. Taken together, the data suggest that syntactic knowledge can facilitate segmentation but that its effect is substantially attenuated if conflicting acoustic cues are encountered before full realization of the syntactic constraint.     

Effects of partial occlusion on perceived surface segregation

Tests are reported of the possibility that local information from contour junctions and from corners of intersecting surfaces is used for perceived surface segregation. Stimuli were two intersecting squares with small disks occluding different parts of the squares. The perceived segregation of the squares from one another decreased as the amount of occlusion of parts of the squares increased. This segregation was less when disks occluded parts of contours than when disks occluded parts entirely inside the squares. Occlusion of parts of contours reduced segregation independently of whether contour junctions or corners were visible or invisible, both where the intersecting surfaces were transparent squares and when they were outlined squares. The present findings show that local information from contour junctions or from corners is not used for surface segregation, and confirm that this segregation is determined by global processes of grouping of areas and of extrapolation of contours.     

Effects of perceived similarity on vicarious emotional conditioning

The present experiment tested the hypothesis that perceived similarity between observers and models enhances vicarious emotional conditioning. Subjects were led to believe that they were either similar to the model in beliefs and attitudes, different from him or were given no information about him. They then underwent a vicarious classical conditioning procedure by witnessing the model express pain reactions in conjunction with a neutral stimulus. Observers developed emotional reactions to the neutral stimulus alone after seeing it paired with modeled pain expressions. Preceived similarity, however, had no significant effect on either the level of vicarious conditioning or the rate of extinction. Subjects who had no information concerning the model displayed stronger autonomic responses to the model's pain expressions, and more frequent and intense autonomic responses to the conditioned stimulus during tests for acquisition and extinction. In correlational analyses, subjects in the latter condition also yielded a significant positive relationship between self-reported arousal and vicarious emotional conditioning. The obtained findings were interpreted in terms of the emotion arousing properties of unfamiliarity.     

Effects of target monitoring on understanding fluent speech

Phoneme monitoring and word monitoring are two experimental tasks that have frequently been used to assess the processing of fluent speech. Each task is purported to provide an “online” measure of the comprehension process, and each requires listeners to pay conscious attention to some aspect or property of the sound structure of the speech signal. The present study is primarily a methodological one directed at the following question: Does the allocation of processing resources for conscious analysis of the sound structure of the speech signal affect ongoing comprehension processes or the ultimate level of understanding achieved for the content of the linguistic message? Our subjects listened to spoken stories. Then, to measure their comprehension, they answered multiple-choice questions about each story. During some stories, they were required to detect a specific phoneme; during other stories, they were required to detect a specific word; during still other stories, they were not required to monitor the utterance for any target. The monitoring results replicated earlier findings showing longer detection latencies for phoneme monitoring than for word monitoring. Somewhat surprisingly, the ancillary phoneme- and word-monitoring tasks did not adversely affect overall comprehension performance. This result undermines the specific criticism that on-line monitoring paradigms of this kind should not be used to study spoken language understanding because these tasks interfere with normal comprehension.     

Effects of symptom presentation order on perceived disease risk

People are quick to perceive meaningful patterns in the co-occurrence of events. We report two studies exploring the effects of streaks in symptom checklists on perceived personal disease risk. In the context of these studies, a streak is a sequence of consecutive items on a list that share the characteristic of being either general or specific. We identify a psychological mechanism underlying the effect of streaks in a list of symptoms and show that the effect of streaks on perceived risk varies with the length of the symptom list. Our findings reveal a tendency to infer meaning from streaks in medical and health decision making. Participants perceived a higher personal risk of having an illness when presented with a checklist in which common symptoms were grouped together than when presented with a checklist in which these same symptoms were separated by rare symptoms. This research demonstrates that something as arbitrary as the order in which symptoms are presented in a checklist can affect perceived risk of disease.     

Effects of gender and topic on speech style

Male and female subjects described nine photographs that had been selected to be of interest to males, females, or both. The resulting speech samples were analyzed for the incidence of five features of word production and five features associated with a women's language (WL) speech style. Results showed that topic (photograph type) determined aspects of speech production. However, there were no gender differences in production or style. Correlations between dependent variables suggested that WL is a consistent set of features. Its use by both female and male speakers may reflect social uncertainty.The authors thank Janis Glenn and Cynthia Johnson for help in collecting and transcribing data.     

Effects of solution viscosity on perceived saltiness and sweetness

Subjects used magnitude estimation to judge the perceived saltiness or sweetness of a series of aqueous solutions containing five suprathreshold concentrations of NaCl or sucrose and thickened with sodium carboxymethylcellulose (CMC). In the first experiment, CMC-H (high viscosity form) was used to thicken a series of sucrose and NaCl solutions to six viscosity levels (1–2,025 centistokes). At the highest viscosity levels, significant decreases occurred in the perceived taste intensity of only the lower concentrations of sucrose and NaCl. A second experiment determined that variations in the quantity of solution sampled from cups did not systematically influence judgments of saltiness when the starting volume was 10 ml. In the third experiment, aqueous solutions containing sucrose or NaCl were thickened with the low (L), medium (M), or high (H) viscosity form of CMC (1–1,296 centistokes). CMC-L-thickened solutions produced little or no suppression of perceived taste intensity, whereas viscous CMC-H solutions produced significant reductions in perceived saltiness and sweetness.     

Shrinkage in the perceived duration of speech and tone by acoustic replacement1

Abstract: To investigate mechanisms for perceiving the duration of an auditory event, an effect of perceptual grouping upon perceived duration was studied psychophysically. In the first experiment, the perceived duration of a spoken word was measured under three conditions of acoustic continuity (i.e., (a) intact, (b) noise‐replaced, and (c) gap‐replaced) as a function of the duration of the target stimulus. Under the noise‐replaced condition, a portion of the target stimulus was physically replaced with a noise burst. Under the gap‐replaced condition, the replacement was made with a gap. The gap‐replacement resulted in a prominent shrinkage of the perceived duration. In the case of noise‐replacement, the amount of shrinkage was moderate but highly significant, although the word employed was perceived to be phonetically intact. Independent of this effect of replacement, the amount of shrinkage was also affected by the physical duration of the target stimulus. The second experiment tested an effect of noise replacement on the perceived duration of a tone burst. In this case, the noise replacement also shrunk the perceived duration of the non‐speech stimulus. This noise‐induced shrinkage could be regarded as being general for the auditory duration. The phenomenon is discussed in relation to a revised model for perceived duration.     

Effects of different motion characteristics on perceived motion in depth

Abstract.— The effect of different velocity characteristics on type of perceived motion were tested with three different stimulus patterns, each representing a certain case of relative motion vectors derived from a vector model for perceived motion in space. The oscilloscope generated patterns, displayed onto a translucent screen, consisted of two dots moving back and forth in their motion paths. The subjects described the perceived motion verbally. The reports were classified into four response categories, i.e. perceived translation in depth, rotation in depth, translation and rotation in depth, and finally, perceived motion in a frontoparallel plane. It was found, first, that no type of relative motion vectors consistently yielded the same distribution of responses for the different velocity conditions. Second, there were no main effects of type of velocity functions (sinusoidal, hyperbolical, and constant) on perceived motion. Third, the position of maximum velocity of the dots affected perceived motion, maximum velocity at the center of the motion path favoring perceived rotation in depth and maximum velocity at the end points of the paths favoring perceived translation in depth. Finally, patterns with continuously repeated motion cycles favored perceived rotation in depth. When the continuity was broken down by pauses at the center and the end points of the motion paths and a small spatial gap at the center of the path, perceived translation in depth was favored.     

Effects of stimulus characteristics and sex of subject on perceived emotion

To determine the influence of pupil size, eye size, sex of stimulus face, and sex of subject on perceived emotion, 308 college students filled in missing features on stimulus faces. Each face was independently rated by two male judges on a continuum of negative to positive emotion. Analysis indicated that sex of subject was the only variable to affect perceived emotion. Women (n = 175) tended to draw faces that were judged to be more positive than faces drawn by men (n = 133).     

The role of phonemes and syllables in the perceived similarity of speech sounds for children

In two experiments, we investigated the factors that influence the perceived similarity of speech sounds at two developmental levels. Kindergartners and second graders were asked to classify nonsense words, which were related by syllable and phoneme correspondences. The results support the existence of a developmental trend toward increased attention to individual phonemic segments. Moreover, one significant factor in determining the perceived similarity of speech sounds appears to be the position of the component correspondences; attention to the beginning of utterances may have developmental priority. An unexpected finding was that the linguistic status of the unit involved in a correspondence (whether it was a syllable or a phoneme) did not seem particularly important. Apparently, the factors which contribute to the perceived similarity of speech sounds in the classification task are not identical to those which underlie performance in explicit segmentation and manipulation tasks, since in the latter sort of task, syllables are more accessible than phonemes for young children. The present task may tap a level of processing that is closer to the one entailed in word recognition and lexical access.