Zeitschrift fur experimentelle Psychologie : Organ der Deutschen Gesellschaft fur Psychologie最新文献

Syntactic ambiguities requiring additional processing usually elicit positive going waveforms with onset latencies between 300 and 600 ms in the event-related potential (ERP). It is still unclear to what extent these components can be viewed either (a) as language specific in nature or (b) as members of the domain-unspecific P300 family of components. The present study investigates this question by means of probability manipulations applied to German sentences with subject-object ambiguities. Non-preferred object-subject (OS) word order requires structural revisions whereas the initially preferred subject-object (SO) word order does not. In the present experiment, the proportions of OS and SO structures were varied across experimental blocks (i.e., .25/.75 vs. .75/.25). The data of 20 participants reveal that ERP components were predominantly influenced when the subjects were explicitly informed about the actual proportions before each single block. In this case an early frontal positive component at about 400 ms and a subsequent posterior positivity were elicited by rare sentence structures irrespective of word order, suggesting that sentence processing was under strategic control. Conversely, participants that were not informed about sentence proportions showed larger positivities to the unpreferred OS sentences. Probability manipulations did not affect this pattern significantly. The data suggest that positivities evoked by syntactic ambiguities respond differently to probability manipulations than the P300 component.

Two experiments, an "active" and a "passive" oddball-paradigm (1. phase) with visual and auditive stimuli were performed. In both, two stimuli with a probability of 0.7/0.3 were presented. The oddball-sequence was replaced in the 2. phase without knowledge of the subjects by a sequence, in which only targets were presented. One group of the subjects has got an active task (counting the rare stimulus), the other group was told to view the stimuli passively. The rare stimulus in the oddball-sequence evoked a larger parietal P3, the auditive stimulation additionally a larger fronto-central N1 and the visual stimulation a larger central P2. The non-expected change into the 2. phase resulted in a reverse of the ERP-lateralization between 150-300 ms: Before the change of the probability the amplitudes were more negative left than right, thereafter more negative right than left. All components and effects in the active task maintained without task, but they decreased. The differences were larger with the visual stimuli. These results suggest, that the passive paradigm could be applied to patients with motoric deficits. From a theoretical point of view these results lead to certain difficulties in the interpretation of the functional importance of the P3. It can be explained better by the context-closure theory (Desmedt, 1980; Verleger, 1988) than by the context-updating theory (Donchin, 1981).

This paper presents an extension of the process dissociation procedure with wordstem completion, allowing the measurement of the stochastic relation between conscious and automatic processes. The conditional probabilities of conscious remembering with and without automatic processes can be determined. The determination of these conditional probabilities succeeds due to the extension of the process dissociation procedure by an indirect wordstem completion test, without referring to the usual assumptions of the dissociation paradigm. For the evaluation of this extended process dissociation procedure a multinomial model is presented which allows the distinction between voluntary and involuntary conscious memory. The model is applied to a study by ourselves and to an experiment published by Toth, Reingold, and Jacoby (1994). The results show that the assumptions of stochastic independence or redundancy as well as exclusivity of conscious and automatic processes are often violated. Two conscious processes with different probabilities of occurrence are found; voluntary and involuntary conscious memory processes.

The features of a visual stimulus are processed in different regions of the visual cortex with no direct axonal connections. Therefore, neurons in the distributed processing areas must be connected in some way to form the physiological substrate of the percept. On the basis of theoretical considerations and animal experiments, it has been proposed that synchronization of neuronal oscillatory firing patterns in the gamma band range (above 30 Hz) might be essential in linking the anatomically distant cell assemblies that represent the various features of the stimulus. The present work reports on three experiments in which the functional relevance of induced gamma band responses were investigated in the human EEG. Using an identical stimulation design, as used in animal studies, it was demonstrated that human induced gamma band responses resembled those reported from intracortical recordings from animals. It was further shown that alpha and gamma band activities differed in temporal characteristics as well as in topographical features, indicating the representation of different cortical functional states. In accordance with previous animal and human experimental findings, a complex moving stimulus was related to a suppression of induced gamma band activity as opposed to a standing complex stimulus.

The lateralized readiness potential as a measure of response tendencies has become a relevant method in experimental psychology within the last decade. This measure delivers information about response preparation long before the overt response. Applying the method of measuring event-related lateralizations (ERL) of the EEG to posterior sites, correlates of processing of spatial information and of response selection have been found. The present study investigated the topographical distribution of asymmetries at the moment of response selection. Response selection was assumed to be reflected in a discrete peak of parietal asymmetries. At this time point two distinct topographies were found that might reflect two different types of information processing. If response selection was based on spatial information a parieto-occipital maximum was found. In contrast, if centrally presented symbolic stimuli were used for response selection, lateralizations were largest at central (motor areas) and at parietal sites.

Event-related potentials (ERPs) of error trials in choice tasks and Go/Nogo tasks are found to be considerably different from the ERPs of the correct trials: In error trial ERPs there is an additional negative (Ne) and an additional positive component (Pe) compared to correct trials. Amplitude and latency variation of both components in different experiments supports the hypothesis that these components reflect different aspects of error processing. The Ne is interpreted as a real-time correlate of error detection, as defined by a mismatch between cognitive representations of the erroneous response and the correct response. The variation of Pe with experimental variables is different from that of the Ne and it also from that of positive components in correct trials, and may therefore reflect an additional aspect of error processing, such as change of response strategies.

Public Speaking is often used to induce anxiety. The emotional stress elicited by the anticipation of delivering a public speech, however, is usually confounded with the mental stress of speech preparation. In order to separate the effects of the emotional from those of the mental stress factor, only half of the subjects were informed about the topic of the speech at the beginning of the anticipation period, whereas the other half were told about the topic at a later time. In a pilot study, the effects of this variation were tested under a condition in which the subjects were told that the speech would be videotaped as an anxiety-provoking and under an emotionally neutral control condition. The main experiment involved an additional condition with a simulated audience designed to intensify the impact of the emotional stress component ("strong anxiety"). Self-reports on present state and cardio-vascular and electrodermal responses were measured in n = 12 subjects in both studies. Knowledge of the speech topic did not affect the subjective anxiety-inducing effects of public speaking. Physiologically arousing effects, however, could be shown without knowledge of the topic only in some variables and only under the "strong anxiety"-provoking condition. In studying public speaking anxiety, confounding with the mental stress of speech preparation should therefore be avoided and a more differentiated interpretation of the physiological effects should be made.

The effects of activation on performance are rarely investigated in older people. An experiment on the effects of psychic activation on free recall of word lists is reported. Participants were healthy women (N = 71) of two age groups (60-69 versus 70-79 years). Psychic activation was induced by intermittent white noise of three intensities (55, 80, 90 dBA). The effects were assessed by self-report inventories concerning mood and performance, by vegetative variables, and by memory tests. The memory tests consisted of free recall tasks under instructions of power and speed. Under the power condition memory performance decreased with activation in both age groups. Free recall performance under speed conditions resulted in inverted U-shaped relations between activation and performance only for the younger sample; the older group was not affected. To explain the results possible mediators are discussed.

This study evaluates a new cognitive rehabilitation therapy for patients after severe head injury. In addition to the standard neurological rehabilitation therapy, one group was trained by the Wiener Determinationsgerät (WDT), a second group was treated by the new program REHACOM, while a third group received only conventional neurological rehabilitation therapy. The three groups each consisted of 12 patients; two groups received 20 sessions of training, each lasting 40 minutes. At the beginning as well as after the therapy a psychological test battery was applied, consisting of HAWIE, TULUC, AACHENER APHASIETEST, and BENTON-Test. They were also tested by a specific neuropsychological battery regarding hemispheric specialization. REHACOM showed significantly higher values on the HAWIE as well as on BENTON-Test than the other two groups. REHACOM also improved in right-hemispheric dimensions while WDT group did not improve in attention. Right-hemispheric training was more effective than attentional stimulation.

In serial choice-reaction-tasks reaction times decrease faster if the sequence of stimuli is structured rather than of being random. Nevertheless, the issue whether this structure-specific improvement of performance is due to the structure of the stimulus- or the response-sequences is controversially discussed. After a review of the corresponding literature, an experiment will be reported which was designed to contribute to this issue. Contrary to the present experiments, which exclusively analysed effects of statistical redundancy, relational patterns are introduced into the sequences of stimuli as well as into the sequences of responses. These relational patterns were varied independently of each other. The data reveal strong effects of the relational patterns in the sequence of responses, whereas the relational patterns of the stimulus sequences were not systematically effective. Furthermore, the data suggest that the relational structure of the response sequences especially favours the formation of "motor-chunks". In the discussion it is emphasized that effects of relational patterns on serial choice reactions are not yet adequately regarded, neither in experimental research nor in the theoretical accounts.