Recognition Failures Research Articles

Trace loss and the recognition failure of unrecalled words

This paper addresses the phenomenon of recognition failure from the perspective of a theory in which recognition and recall are assumed to involve independent retrieval processes. However, even given independent retrieval, measures of recognition and recall success will covary if any traces are lost from storage, simply because such traces are unavailable for any memory test. In support of the theory, rote learning produced higher covariation between recognition and recall (i.e., fewer recognition failures) than did meaningful elaboration during study. Further, recognition and recall were approximately independent of each other with meaningful elaboration and imagery encoding, regardless of whether the latter involved interactive or separate images. The results of three experiments are discussed in terms of the present “vandal” theory and other theories of recognition failure.

Cognate Recognition across Dialects

Abstract Synchronic cognates are subject to recognition failures across dialects, if processing through different determining features leads to phonemic misclassification. But phonemes common to all dialects can also cause recognition failures if they differ in function. These findings support a theoretical distinction between phone, phoneme, and morpho phone, the last being an essential unit for the cognate reconstruction necessary for error-free processing of unfamiliar dialects.

A method for evaluating computer programs for electrocardiographic interpretation. I. Application to the experimental IBM program of 1971.

A method for evaluating computer programs for electrocardiographic interpretation is described. This method allows a clinician to judge the usefulness of a program for his specific setting and needs. The method requires a significant proportion and variety of abnormal tracings, the application of specific fixed criteria, and the separation of disagreements between the computer program and the clinician into those resulting from criteria differences and those resulting from programming errors, viz., pattern recognition failures, mismeasurements, and/or deficient program logic. When applied to the experimental IBM program 1971, staff cardiologists found essential agreement with the program's results in 76% of 1150 unselected tracings. Clinically significant disagreements based strictly on the application of different criteria occurred in 20% of the tracings, whereas disagreements based on program errors were found in only 4%. Although this program requires some system of human overview and quality checking, its potential for clinical implementation is worthy of consideration.

A method for evaluating computer programs for electrocardiographic interpretation. II. Application to version D of the PHS program and the Mayo Clinic program of 1968.

A previously described method for evaluating computer programs for electrocardiographic (ECG) interpretation was applied to Version D of the Public Health Service (PHS) program and to the Mayo Clinic program of 1968. Staff cardiologists found agreement with the results of the PHS program in 45.5% of 1150 unselected tracings. Clinically significant disagreements based strictly on application of different criteria occurred in 29%, while disagreements based on program errors were found in 25.5%. The corresponding results for the Mayo Clinic program are: agreement in 47%, disagreements due to criteria differences in 30.9%, and disagreements due to program errors in 22.1%. Both programs had serious deficiencies, particularly in the diagnostic categories of myocardial infarction and cardiac arrhythmias. PHS program errors resulted primarily from mismeasurements and deficient program logic, while Mayo Clinic program errors more frequently resulted from pattern recognition failures. Neither program appears suitable for routine clinical use at the present time.

Arrhythmia detection by digital computer

A digital computer program has been described for automatic interpretation of the most common arrhythmias, which can be recognized in 6- and 10-second ECG recordings. A decision-tree method was applied for classification. The logic attempted to compensate for possible measurement or pattern recognition failures. The program consisting of approximately 1700 FORTRAN instructions, was developed using a learning file of various arrhythmias and was tested against independent new records. Correct diagnosis of sinus rhythm without atrioventricular block was obtained in 86–97% of the cases dependent upon the material analyzed in 6- as well as 10-second recordings. Atrial fibrillation was correctly recognized in 82 and 92%, respectively, of short and long records. Eighty-seven percent of ventricular extrasystoles and aberrant beats were found correctly with a relatively low incidence of false positives, which compares favorably with results published by others. The limitations of the program and of arrhythmia detection in general, using short strips recorded with surface ECG leads, are discussed.

The IBM 1975 Optical Page Reader Part II: Video Thresholding System

An adaptive video thresholding system is used in the Page Reader to minimize recognition failures due to contrast and line width variation. The main threshold operator is a linear function of the average contrast over a specified area. Adjustments to the threshold level are made by circuits that compute the average line widthin a character, and by circuits that filter out spatial noise in the vicinity of the character. The different types of print quality distortions predominant in typewriter printing and business machine printing are handled by switching between different sets of threshold operators.

Memory and the memory-monitoring process

Two experiments are reported that attempt to evaluate whether people can accurately monitor the contents of their memories when they are unable to retrieve those memories. Both experiments used recently learned paired-associates as memory materials; S s were asked to predict which unrecalled response items they would be able to recognize by referring to their feelings of knowing about the missing items. The results show that the Ss were able to make relatively accurate predictions about recognition failures and successes. The general process revealed by the findings (called the memory-monitoring or MEMO process) is discussed in relation to recall and recognition thresholds.