Sensory Discrimination Task Research Articles

Calibration of sensory and cognitive judgments: A single model for both

In a recent issue of this journal, Winman and Juslin (34, 135-148, 1993) present a model of the calibration of subjective probability judgments for sensory discrimination tasks. They claim that the model predicts a pervasive underconfidence bias observed in such tasks, and present evidence from a training experiment that they interpret as supporting the notion that different models are needed to describe judgment of confidence in sensory and in cognitive tasks. The model is actually part of the more comprehensive decision variable partition model of subjective probability calibration that was originally proposed in Ferrell and McGoey (Organizational Behavior and Human Performance, 26, 32-53, 1980). The characteristics of the model are described and it is demonstrated that the model does not predict underconfidence, that it is fully compatible with the overconfidence frequently found in calibration studies with cognitive tasks, and that it well represents experimental results from such studies. It is concluded that only a single model is needed for both types of task.

Spatial cueing, sensory gating and selective response preparation: an ERP study on visuo-spatial orienting

Event-related brain potentials (ERPs) were recorded in a visuo-spatial attention task where the position of an imperative stimulus was indicated either validly or invalidly by a central arrow (trial-by-trial cueing). Subjects had to perform choice RT tasks with the response being dependent either on the identity of the target stimulus or on its position. When target identity was relevant for response selection, validly cued stimuli elicited amplitude enhancements of the early, sensory-evoked P1 and N1 components at lateral posterior sites. The N1 validity effect was limited to scalp sites ipsilateral to the visual field of stimulus presentation. Although these effects were found only when the sensory discrimination task was considerably difficult, they are in line with models assuming that modulations of sensory-perceptual processing (“sensory gating”) are induced by spatial cueing. However, when target location was response-relevant, N1 amplitude enhancements were consistently elicited by invalidly cued letters. CNV and LRP measures indicated that the arrow elicited response-related processing in the cue-target interval. Such processes occurred even when the cue contained no information about an upcoming response. Two consecutive lateralization phases were distinguishable in the LRP, with experimentally induced response assignments becoming effective only during the second phase.

Discrimination performance of single neurons: rate and temporal-pattern information

1. A new method of measuring the performance of neurons in sensory discrimination tasks was developed and then applied to single-neuron responses recorded in the auditory nerve of chinchilla and in the striate visual cortex of cat. 2. Most previous methods of measuring discrimination performance have employed decision rules that involve comparing the total counts of action potentials (spikes) produced by two different stimuli. Such measures ignore response pattern and hence may not reflect all the information transmitted by a neuron. The proposed method attempts to measure all (or most) of the transmitted information by constructing descriptive models of the neuron's response to each stimulus in the discrimination experiment; these descriptive models consist of measured probability distributions of the spike counts in small time bins. The measured probability distributions are then used to define an optimal decision rule (an ideal observer) for discriminating the two stimuli. Finally, discrimination performance is measured by applying this decision rule to novel presentations of the same two stimuli. 3. Intensity and temporal-phase discrimination were measured for three neurons in the auditory nerve of chinchilla. The discrimination stimuli were low-frequency pure tones of 70-ms duration. Intensity thresholds were found to be 5-20 dB lower at low intensities using the new pattern method compared with the traditional counting method. The pattern method led to better performance because it utilized both rate and temporal pattern information. Phase discrimination performance using the counting method was at chance because the average spike rate did not change with phase. On the other hand, using the pattern method, phase discrimination thresholds were found to decrease with intensity, often reaching values equivalent to 30-40 microseconds of temporal offset. These thresholds are as good as or better than behavioral thresholds in chinchilla. 4. Contrast and temporal-phase discrimination were measured for three neurons in the striate visual cortex of cat. The discrimination stimuli were drifting sine-wave gratings of 100- to 160-ms duration. Contrast discrimination functions measured by the pattern method and the counting method were found to be essentially identical. Phase discrimination using the counting method was at chance. However, using the pattern method, phase thresholds were found to decrease with contrast, reaching values equivalent to 7 ms of temporal offset for the two simple cells. 5. Our results suggest that temporal response pattern carries substantial information for intensity and phase discrimination in the auditory nerve and for phase discrimination in the striate visual cortex.(ABSTRACT TRUNCATED AT 400 WORDS)

Spatio-Temporal Cortical Patterns Evoked in Monkeys by a Discrimination Task

Abstract The primary experimental objective of this work was to demonstrate localization and temporal sequencing of the functional steps carried out by nonhuman primate subjects during performance of a sensory discrimination task, i.e., to identify the locale and sequence of activation of regions that participate in sensory discrimination, stimulus classification, and response preparation. Multivariate statistical procedures were applied to evoked transcortical recordings to identify the location and order of occurrence of signals that are effective in discriminating task conditions and parameters. (1) Sensory discrimination, (2) stimulus classification, and (3) response preparation occurred in the expected sequence. Information that enabled discrimination using these procedures was distributed widely across the cortex; however, the maximum information content was localized to striate and prestriate cortex, anterior inferior parietal cortex, and temporal and premotor cortex, respectively. This work provides a perspective on brain mechanisms responsible for cognition and demonstrates a set of powerful multivariate analytic tools for functional mapping, i.e., identifying the location and sequencing of cognitive functions.

Lesions of parvocellular subdivisions of the hypothalamic paraventricular nucleus alter open field behavior and acquisition of sensory and spatial discrimination

Rats with ibotenic acid (IBO) lesions of the hypothalamic paraventricular nucleus (PVN) were compared with operated control animals over a battery of tests designed to assess memory- and arousal-related behavioral processes. At the dose employed in these experiments, ibotenic acid selectively destroys parvocellular elements of the PVN, leaving magnocellular subdivisions relatively intact, allowing for experimental dissection of the influence of parvocellular and magnocellular PVN neuronal populations on the behavioral parameters measured. IBO-treated rats showed a greater incidence ofrearing behavior and exhibited greater levels of total and central ambulation in an open field than control rats. Acquisition of both sensory and spatial reward contingencies were retarded in the IBO-lesion group; however, no differences were evident between IBO-treated and control groups in an approach-avoidance test, nor in the ability to perform the spatial and sensory discrimination tasks to a criterion level of accuracy. Histological examination verified that bilateral IBO lesion destroyed parvocellular elements of the PVN, while sparing the majority of magnecellular neurons. Results suggest that parvocellular PVN lesion alter behavioral performance via interactions with physiological systems overning arousal level.

NEURONAL ACTIVITY IN THE HUMAN MEDIAL TEMPORAL LOBE DURING RECOGNITION MEMORY

Human medial temporal lobe neuronal activity and event-related potentials were recorded during the following behaviours: contextual recognition of words and faces, semantic discrimination of nonwords from words, and discrimination of stimulus classes based on perceptual attributes. Three distinct classes of behavioural correlates of unit activity were demonstrated by visual inspection of peristimulus histograms and by nonparametric statistics: (1) neuronal excitation during a keypress related to the subject's choice; (2) specific and nonspecific excitation to words; (3) excitation or inhibition to rare stimuli in a sensory discrimination task. Responses specifically to familiar (as opposed to unfamiliar) words or faces, or to tasks requiring recent memory per se were never seen. Keypress excitation was relatively common (32/76 units) and occurred regardless of whether the keypress target was a repeated or nonrepeated word, or the task required recent or remote semantic memory. In a more complex recognition task utilizing two responses and an imperative cue for the patient's response, units with prior keypress excitation failed to generate the response. This suggests that keypress excitation is not strictly tied either to response choice or to generation. The onset latencies and temporal relationship to event-related potentials of the nonspecific and specific excitation to words and the excitation to rare stimuli suggest that they represent contextual encoding of stimuli. Similar evidence suggests that the inhibition to rare stimuli represents inhibitory processes terminating contextual encoding. Thus human medial temporal lobe neurons seem to contribute information during successive stages of a cognitive stimulus-response task: contextual encoding, closure and response-selection.

Subclasses of event-related potentials: response-locked and stimulus-locked components.

The manner in which information processing proceeds and the relationship of the components of the cerebral "event-related" evoked potential (ERP) to the presumed stages of such processing are uncertain. We studied these relationships by recording long-latency evoked potentials from the scalp in conjunction with electromyographic (EMG) recording in experiments in which subjects were required to make a motor response to one of two auditory stimuli. The cerebral response was averaged from the tone onset (stimulus-synchronized average) and the EMG onset (response-synchronized average). We found that N2 was quite distinct from the other ERP components (P165 and P3) in its lack of a close relationship to the motor response required in a sensory discrimination task. This suggests that, contrary to general belief, the P165-N2-P3 components of the ERP are not generated in an orderly sequence, but rather are generated separately and are possibly related to different cognitive processes.

Differential effects of posterior septal lesions on dispositional and representational memory.

A distinction between two classes of memory has been made in terms of the sensory availability of cues at the time of making discriminations which are influenced by past experience. Three tasks objectively defining this distinction were learned in a T-maze by three groups of rats: a delayed nonmatching-to-sample (DNMTS) which depends on representational memory; a simple sensory discrimination (SD) which depends on dispositional memory; and a more difficult discrimination, which also depends on dispositional memory, called the simultaneous conditional discrimination (SCD). The DNMTS and SD tasks were acquired quickly; the SCD task took many more trials. Posterior septal lesions impaired DNMTS performance but had no effect on retention of either the conditional discrimination (SCD) or the simple sensory discrimination (SD) tasks which depend on dispositional memory. The present study provides further evidence that dispositional and representational memory systems have at least partially distinct anatomical substrates in the brain and that it is the representational and not the conditional aspects of the DNMTS task that are impaired by the septal lesions.

Entorhinal and Septal Inputs Differentially Control Sensory-Evoked Responses in the Rat Dentate Gyrus

Averaged sensory-evoked potentials were recorded from the outer molecular layer of the dentate gyrus in naïve rats and in rats conditioned to respond in the presence of an auditory stimulus. Two components (negative peaks) of the potentials were functionally distinguished in terms of responsiveness to unique or conditioned auditory stimuli. Each component was independently generated by a separate input pathway to the dentate gyrus: The perforant path provided an "insignificance" or "unexpected" feature of the sensory stimulus when appropriate, and the septum controlled the development of a second component as a function of the behavioral significance of the stimulus during the acquisition of auditory discrimination behavior. A reciprocal relationship between the peak amplitudes of both components of the average evoked potentials dependent on the relative behavioral significance of the sensory stimulus was observed in all animals during extinction and reconditioning of the sensory discrimination task. The findings indicate that the entorhinal and septal projections to the dentate granule cells are activated differentially by sensory stimuli as a function of their acquired behavioral significance to the animal.

A Brain Event Related to the Making of a Sensory Discrimination

Event-related potentials associated with detected targets in a vigilance task were analyzed in two ways: (i) by sorting the potentials in terms of sequential reaction time bins of 50 milliseconds and (ii) by examining the single trial waveforms. A negative component (N2) covaried in latency with reaction time. These results support the hypothesis that N2 reflects a decision process which controls behavioral responses in sensory discrimination tasks.

Satiation effects in continuous reinforcement and successive sensory discrimination situations

Nine rats with forebrain and hypothalamic electrodes were trained on a sensory discrimination task with S–intervals ranging from 12 to 60 sec. Their performance during a 48-h discrimination session was compared to that during a 48-h CRF session. The response measures showed that forebrain Ss performed more poorly than hypothalamic Ss during the discrimination session, even though performance of forebrain animals was better in the discrimination session than in the CRF session when time until the first 5-min pause was considered. In the CRF session, there was no difference in performance of forebrain and hypothalamic animals, except in terms of time till first pause. The results are discussed in terms of cumulative effects of forebrain and hypothalamic ESB.

Articulatory effectiveness, stimulability, and children's performances on perceptual and memory tasks.

The performances of 70 children, 35 from kindergarten and 35 from first grade, were studied on four auditory measures and one oral sensory discrimination task. Each group of 35 children included seven subjects with superior articulation, seven with deviant articulation and poor speech sound stimulability, seven with deviant articulation and good stimulability, seven with articulation defects and poor stimulability, and seven with articulation defects and good stimulability. Performances on a speech sound stimulability task were not found to be related to performances on any of the auditory measures and only slightly to the oral sensory task. Superior articulators had significantly better scores than the deviant and defectives on the oral sensory discrimination task, but scores on the auditory tasks were not significantly different. Comparison of the performances of /s/ and /r/ defectives revealed the latter group to be inferior on some auditory tasks compared with the superior articulators. Some relationships were found between subject performances on certain auditory tasks.

The relative ability of aphasic persons to judge the duration and intensity of pure tones.

This investigation was undertaken to determine the relative ability of aphasic individuals to perform specific sensory discrimination tasks. The study sought to determine if aphasics differ from no...

Mediating processes in choice reaction tasks

A ten-choice reaction time (RT) experiment employing two subjects ( Ss) is described. One S completed the experiment with stimulus factors held constant and response factors balanced out, while the other S completed the experiment under the reverse conditions. The results give strong evidence of an inverted U shaped relationship between RT and position of the individual stimuli and responses within their respective arrays. The findings were interpreted as supporting the view that performance in choice RT tasks is mediated by internalised representations of the stimulus and response arrays. The properties of these internalised representations are consistent with those found in other sensory discrimination tasks, and it is argued that they account for the basic form of the relationship between RT and number of stimulus alternatives.