PubMed · 14776178
The source did not provide an abstract. Follow the original record for more information.
Explore connections, maps & timelines
C L HULL. 1950. Simple qualitative discrimination learning.. https://doi.org/10.1037/h0062099
Cite the original work for its findings. Save a collection to share your selection of sources.
KEEP EXPLORING
The two-point discrimination (2PD) test is the most frequently used test for the assessment of the sensory outcome after nerve repair. Here we focus on factors which explain the enormous and implausible variability in reported 2PD levels after nerve repair. We conclude that the 2PD testing technique is not at all standardized and that its use as the sole test for tactile gnosis recovery should be seriously questioned. Reports of 2PD results should always be accompanied by a detailed description of how the test was performed, especially with reference to the pressure applied and the testing protocol.
We studied perceptual learning in motion discrimination when the brain's middle temporal area (MT/V5) was functionally suppressed. This was achieved by using the "paired-dots" motion stimulus where the two dots in a pair always move in counter-phase over a short distance [J. Neurosci. 14 (1994) 7357]. The motion directional signal of the stimulus is therefore always 0 on average. As a result, this stimulus activates MT in Rhesus monkeys no more than flicker noise does [J. Neurosci. 14 (1994) 7367]. We added a new manipulation to eliminate the Glass pattern in the original stimulus that would have otherwise provided a static orientation cue. Two such new motion stimuli were presented sequentially, in a 2AFC task. Subjects decided if the global motion-axis of the stimuli changed clockwise or counter-clockwise. When the task difficulty was set at 60% correct, none of the subjects could learn with feedback, even though their performance was well above chance. However, when the task difficulty was set instead at 70% correct, a new group of subjects was able to learn. Hence, learning motion discrimination was possible when MT was presumably eliminated.
In character string recognition integrating segmentation and classification, high classification accuracy and resistance to noncharacters are desired to the underlying classifier. In a previous evaluation study, the modified quadratic discriminant function (MQDF) proposed by Kimura et al. was shown to be superior in noncharacter resistance but inferior in classification accuracy to neural networks. This paper proposes a discriminative learning algorithm to optimize the parameters of MQDF with aim to improve the classification accuracy while preserving the superior noncharacter resistance. We refer to the resulting classifier as discriminative learning QDF (DLQDF). The parameters of DLQDF adhere to the structure of MQDF under the Gaussian density assumption and are optimized under the minimum classification error (MCE) criterion. The promise of DLQDF is justified in handwritten digit recognition and numeral string recognition, where the performance of DLQDF is comparable to or superior to that of neural classifiers. The results are also competitive to the best ones reported in the literature.