The efficiency of holistic template matching in the recognition of unconstrained handwritten digits

Summary Psychological evidence suggests that simple visual patterns can be recognized by the use of internal representations as holistic templates, but the efficiency of holistic template matching in the recognition of real-life patterns, such as handwritten characters, has been doubted. To clarify this issue, we measured the efficiency of holistic template matching in machine recognition of totally unconstrained handwritten digits. Our learning and recognition algorithm was simple; no previous knowledge of handwritten digits was presupposed, and preprocessing was limited to Gaussian smoothing and normalization with respect to position, size, and orientation. For patterns presented in a known orientation, recognition rates were .69, .77, and .88, respectively, when about 5, 10, or 50 templates had been learned for each type of digit. For patterns presented in unknown orientations, recognition rates were slightly lower. High levels of reliability could be attained by the discounting of classifications based on weak evidence. Apparently, in high reliability recognition, holistic template matching can be used as a first operation by which recognition is achieved for most of the handwritten digits that are seen in real life.     

The dependency of the validity of information integration on cognitive variables and the judgement task

Simple models of information integration focus essentially on the combination of the components of information. This research investigated whether cognitive variables in the constructs of intelligence and cognitive complexity, as well as concentration, could predict the conditions of simple models of judgement. As additional predictors, the two qualitative variables [type of information] and [experience of the judgement task] were introduced. Subjects judged three types of stimuli using the pair comparison method. The conditions of the judgement models were analysed in the framework of the conjoint-measurement approach. Five different regression functions provided mediocre approximations to the 10 conditions of the models, when the 9 cognitive variables were considered. The small differences did not favour one of the five functions. The introduction of the qualitative predictor [type of information] improved the estimations more clearly than the experience of the judgement task, the effect of which was small.     

Effects of Gender on Costly Punishment

Women are often believed to be more cooperative and less egoistic than men. In the present study, we examined whether people punish women for failing to live up to these benevolent gender stereotypes. Participants played a prisoner's dilemma game with female and male partners who either cooperated or defected. Participants were offered a costly punishment option. They could spend money to decrease the payment of their partners. In Experiment 1, participants spent more money to punish the defection of female in comparison to male partners, but this effect of partner gender on punishment was indirect rather than direct: Participants were more likely to cooperate with female partners than with male partners, which gave them more opportunity for moralistic punishment. In Experiments 2 and 3, we examined the effects of the participants' own gender on cooperation and punishment of female and male partners. Female participants cooperated more with female partners than with male partners while male participants treated female and male partners equally. We conclude that the effect of facial gender on punishment are indirect rather than direct. The results also showed that women, in contrast to men, tended to make decisions that can be considered more social and less rational from an economic point of view, consistent with social‐role theory and evolutionary accounts. Copyright © 2017 John Wiley & Sons, Ltd.     

The costs of target prioritization and the external requirements for using a recall-to-reject strategy in memory exclusion tasks: a meta-analysis

In event-related potential (ERP) studies, the left-parietal old/new effect is commonly considered as a neural correlate of recollection. In memory exclusion tasks, the effect is usually observed when the targeted information is identified, but it is not necessarily present when studied items are rejected as nontargets. Interestingly, both the presence and the absence of such old/new effects to nontargets have been regarded as indicator for strategic retrieval. We reviewed previous ERP studies using memory exclusion tasks to analyze the reaction time (RT) pattern in such studies, as well as the influence of task difficulty on the occurrence of nontarget retrieval. We identified 44 test conditions, reported in 24 studies, and subjected the behavioral data to a meta-analysis. The RTs to correctly rejected new items were shorter than the RTs to hits, in particular in studies that required differentiating conceptual or perceptual information at test. When the retrieval of target information was prioritized, RTs to nontargets were delayed relative to targets. Without such prioritization, no such RT differences were observed. For test conditions with nontarget retrieval, the retrieval accuracy was poorer compared with test conditions without such retrieval. The findings support previous studies that claimed that nontarget retrieval becomes more likely when target retrieval is difficult, but the strong overlap in task difficulty between conditions with and without nontarget retrieval indicates that other, partly yet to-be-identified factors contribute to the occurrence of nontarget retrieval as well.     

Semantic priming by irrelevant speech

To-be-ignored, task-irrelevant speech disrupts serial recall performance relative to a quiet control condition. In most studies, the content of the auditory distractors had no effect on their disruptive potential, one’s own name being one of the few exceptions. There are two possible explanations of this pattern: (1) Semantic features of the irrelevant speech are usually not processed, except for highly relevant auditory distractors, or (2) semantic processing of the irrelevant speech always occurs, but usually does not affect serial recall performance. To test these explanations, we presented to-be-ignored auditory distractor words drawn from different categories while participants memorized visual targets for serial recall. Afterwards, participants were invited to what they believed to be an unrelated norming study, in which they were required to spontaneously produce words from the categories from which the auditory distractor words were drawn. Previously ignored words were produced with a higher probability than words from a parallel, nonpresented set, demonstrating that features of to-be-ignored, task-irrelevant speech that do not interfere with immediate serial recall performance are nevertheless processed semantically and may have substantial effects on subsequent behavior.     

Auditory location negative priming: A case of feature mismatch

In an auditory four-alternative forced choice localization task, participants had to localize one of two simultaneously presented sounds while ignoring the location of the second sound. Negative priming—that is, sloweddown responses to a location that had to be ignored in the previous trial—was found only when the sound at the repeated location changed between prime and probe. There was also no increase in prime response errors to the probes of ignored repetition trials. These findings allow for the conclusion that auditory location priming is caused by feature mismatch only and that other mechanisms, such as inhibition of ignored locations or episodic retrieval of transfer-inappropriate prime information, do not play a role. The research reported in this article was supported by a grant from the Deutsche Forschungsgemeinschaft (Ma 2610/2-1).     

Rigorous results,cross-model justification,and the transfer of empirical warrant: the case of many-body models in physics

This paper argues that a successful philosophical analysis of models and simulations must accommodate an account of mathematically rigorous results. Such rigorous results may be thought of as genuinely model-specific contributions, which can neither be deduced from fundamental theory nor inferred from empirical data. Rigorous results provide new indirect ways of assessing the success of models and simulations and are crucial to understanding the connections between different models. This is most obvious in cases where rigorous results map different models on to one another. Not only does this put constraints on the extent to which performance in specific empirical contexts may be regarded as the main touchstone of success in scientific modelling, it also allows for the transfer of warrant across different models. Mathematically rigorous results can thus come to be seen as not only strengthening the cohesion between scientific strategies of modelling and simulation, but also as offering new ways of indirect confirmation.     

INDEFENSIBLE MIDDLE GROUND FOR LOCAL REDUCTIONISM ABOUT TESTIMONY

Local reductionism purports to defend a middle ground in the debate about the epistemic status of testimony-based beliefs. It does so by acknowledging the practical ineliminability of testimony as a source of knowledge, while insisting that such an acknowledgment need not entail a default-acceptance view, according to which there exists an irreducible warrant for accepting testimony. The present paper argues that local reductionism is unsuccessful in its attempt to steer a middle path between reductionism and anti-reductionism about testimonial justification. In particular, it challenges local reductionism 'from within', without appealing to anti-reductionist intuitions. By offering novel arguments to the effect that local reductionism fails by its own standards, the present paper considerably strengthens the case against this version of reductionism. Local reductionism, it is argued, fails for three main reasons. First, it cannot account for the rationality of testimonial rejection in paradigmatic cases, even though the possibility of rational rejection is thought to be of central justificatory importance. Second, it does not provide a sufficiently distinct non-testimonial basis to which testimonial justification can be successfully reduced. Finally, local reductionism is shown to be an intrinsically unstable position, in danger of collapsing into full-fledged 'credulism' of the kind historically associated with Thomas Reid.     

Manipulative success and the unreal

In its original form due to Ian Hacking, entity realism postulates a criterion of manipulative success which replaces explanatory virtue as the criterion of justified scientific belief. The article analyses the foundations on which this postulate rests and identifies the conditions on which one can derive a form of entity realism from it. It then develops in detail an extensive class of counterexamples, drawing on the notion of quasi-particles in condensed matter physics. While the phenomena associated with quasi-particles pass the entity realist's criterion of manipulative success, quasi-particles themselves are illusions, and can be seen to be so even on the basis of the largely non-theoretical “home truths” that one must be ready to admit as background knowledge. Hence, Hacking's entity realism is shown to be incoherent.     

Measuring consciousness: relating behavioural and neurophysiological approaches

The resurgent science of consciousness has been accompanied by a recent emphasis on the problem of measurement. Having dependable measures of consciousness is essential both for mapping experimental evidence to theory and for designing perspicuous experiments. Here, we review a series of behavioural and brain-based measures, assessing their ability to track graded consciousness and clarifying how they relate to each other by showing what theories are presupposed by each. We identify possible and actual conflicts among measures that can stimulate new experiments, and we conclude that measures must prove themselves by iteratively building knowledge in the context of theoretical frameworks. Advances in measuring consciousness have implications for basic cognitive neuroscience, for comparative studies of consciousness and for clinical applications.