Language and connectionism: the developing interface

After a difficult initial period in which connectionism was perceived as either irrelevant or antithetical to linguistic theory, connectionist concepts are now beginning to be brought to bear on basic issues concerning the structure, acquisition, and processing of language, both normal and disordered. This article describes some potential points of further contact between connectionism and linguistic theory. I consider how connectionist concepts may be relevant to issues concerning the representation of linguistic knowledge; the role of a priori constraints on acquisition; and the poverty of the stimulus argument. I then discuss whether these models contribute to the development of explanatory theories of language.     

Challenging the ability intuition: From personal to extended to distributed belief-forming processes

Much of what we know results from information sources on which we epistemically rely. This fact about epistemic reliance, however, stands in tension with a very powerful intuition governing knowledge, an intuition that Pritchard (e.g., 2010) has termed the “ability intuition,” the idea that a believer's “reliable cognitive faculties are the most salient part of the total set of causal factors that give rise to [their] believing the truth” (Vaesen, 2011, p. 518; compare Greco, 2003; 2009; 2010). In this paper I suggest that this tension may indeed be ineliminable. I proceed by canvassing some representative attempts to reconcile epistemic reliance and the ability intuition. In doing so, I suggest that all of these attempts founder on one or the other of two elements of what I've previously described (Shieber, 2013, 2015) as a “personalist presumption” in discussions of social epistemology: an excessive focus on (i) reliability filters within the persons who are the recipients of information or (ii) on reliable truth-tracking and -conveying abilities in the persons who are the transmitters of information. In conclusion, I suggest how best to resolve the tension: by abandoning the ability intuition.     

The impact of feedback frequency on learning and task performance: Challenging the “more is better” assumption

Previous research on feedback frequency suggests that more frequent feedback improves learning and task performance (Salmoni, Schmidt, & Walter, 1984). Drawing from resource allocation theory (Kanfer & Ackerman, 1989), we challenge the “more is better” assumption and propose that frequent feedback can overwhelm an individual’s cognitive resource capacity, thus reducing task effort and producing an inverted-U relationship with learning and performance over time. We then propose that positive and negative affective states will moderate the inverted-U relationship between feedback frequency and task performance. We test these propositions in an experimental study where the frequency of task feedback is manipulated. Results show that feedback frequency exhibits an inverted-U relationship with task performance, and this relationship is mediated by task effort. This curvilinear relationship is then moderated by individual’s positive affective state.     

Representations without rules,connectionism and the syntactic argument

Terry Horgan and John Tienson have suggested that connectionism might provide a framework within which to articulate a theory of cognition according to which there are mental representations without rules (RWR) (Horgan and Tienson 1988, 1989, 1991, 1992). In essence, RWR states that cognition involves representations in a language of thought, but that these representations are not manipulated by the sort of rules that have traditionally been posited. In the development of RWR, Horgan and Tienson attempt to forestall a particular line of criticism, theSyntactic Argument, which would show RWR to be inconsistent with connectionism. In essence, the argument claims that the node-level rules of connectionist networks, along with the semantic interpretations assigned to patterns of activation, serve to determine a set of representation-level rules incompatible with the RWR conception of cognition. The present paper argues that the Syntactic Argument can be made to show that RWR is inconsistent with connectionism.The present paper has been improved by conversations with Terry Horgan and John Tienson. Thanks are also due to Gary Fuller, John Heil, Terry Horgan and Bob Stecker for comments on earlier drafts of this paper.     

Exploring the breadth of the top-down representations that control attentional disengagement

Recent studies attribute top-down control as the primary determinant of the speed with which attention can be disengaged from an object, and disengagement has received increased interest as a means to distinguish between top-down and bottom-up accounts of attention capture. We present the results of three experiments exploring the breadth of the representations that delay attentional disengagement based on top-down goals. Experiments 1 and 2 examined whether objects similar to an observers' search target, but differing in luminance and hue, delayed the reallocation of attention in a search paradigm designed to isolate disengagement time. Experiment 3 explored whether the representations that delay disengagement are based on absolute similarity with the search target, or are tuned based on target/nontarget relationships. These three studies confirmed the role of top-down goals in automatically contributing to dwell times and revealed that the representations that underlie disengagement effects are broad (automatically delaying disengagement for items similar, but not identical to, the search target). In some cases, attention sets appeared to be graded in nature, but in others target–distractor relationships influenced the degree to which an irrelevant item held attention. Implications for theories of attention capture and potential functional significance of these automatic effects are discussed.     

Implicit and explicit memory: Implications for the symbol-manipulation versus connectionism controversy

Summary This paper deals with three issues. First, the symbol-manipulation versus connectionism controversy is discussed briefly. Both can be seen as formalisms or languages for describing human behavior, but it is argued that these languages are not fully equivalent. The importance of the ability of connectionist models to learn autonomously, often underestimated, allows them to perform tasks that may defy formalization. Secondly, the structural and functional characteristics of a new connectionist learning model are presented. It avoids some of the psychological and biological implausibilities of currently popular models and solves some of the problems and shortcomings of these models. Thirdly, some simulation results are presented. The model successfully simulates the dissociation between explicit and implicit memory performance as found in patients with anterograde amnesia. This simulation provides an example of the advantages of using the connectionist language in the domain of memory psychology. It is also a good example of how psychological evidence can be used to improve connectionist models.The second author is supported by the Dutch Organization for Scientific Research (NWO)     

The role of sparsely distributed representations in familiarity recognition of verbal and olfactory materials

We present the generalized signal detection theory (GSDT), where familiarity is described by a sparse binomial distribution of binary node activity rather than by normal distribution of familiarity. Items are presented in a distributed representation, where each node receives either noise only, or signal and noise. An old response (i.e., a “yes” response) is made if at least one node receives signal plus noise that is larger than the activation threshold, and item variability is determined by the distribution of activated nodes as the threshold is varied. A distinct representation leads to better performance and a lower ratio of new to old item variability, than a more distributed and less distinct representations. Here we apply the GSDT to empirical data on verbal and olfactory memory and suggest that verbal memory relies on a distinct neural item representation, whereas olfactory memory has a fuzzy neural representation leading to poorer memory and inducing a larger ratio of new to old item variability.     

Political deviance and the assumption of responsibility

Political deviance is a broad concept, less frequently invoked than the concept of political crime. It can be perpetrated by those in power, in the name of the state or individually, or by those struggling to effect social change. The focus on the latter brings to the fore two types of movements, terrorism and civil disobedience, with their partial similarities in goals and their apparent dissimilarities in the means employed to achieve these goals. Both groups tend to assume responsibility for their acts, although for different reasons. The assumption of responsibility itself is a clouded and sometimes ambiguous notion, both linguistically and in its ramifications in the political arena.     

Contractarianism and the assumption of mutual unconcern

I have benefited from suggestions from Grant Brown, Morry Lipson, and an anonymous referee for this journal.     

Sex typing and gender ideology: two facets of the individual's gender psychology that go together

Three studies tested the hypothesis that gendered personality dispositions are related to gender attitudes and gender discriminatory behaviors. In Study 1, sex-typed individuals were more likely than androgynous, undifferentiated, and cross-sex-typed individuals to accept gender rules designating culturally appropriate behavior for men and women. In Study 2, sex-typed individuals were particularly likely to pay attention spontaneously to the sex of job applicants and then to devalue the interview performances of women. In Study 3, only sex-typed men tended to endorse sexist language. As expected, sex typing and gender ideology go together. This relation between two facets of the individual's gender psychology indicates that some sex role inventories may tap more than expressivity and instrumentality.     

Sensory and conceptual representations in memory: Motor images that cannot be imaged

Summary The paper argues for a distinction between sensory-and conceptual-information storage in the human information-processing system. Conceptual information is characterized as meaningful and symbolic, while sensory information may exist in modality-bound form. Furthermore, it is assumed that sensory information does not contribute to conscious remembering and can be used only in data-driven process reptitions, which can be accompanied by a kind of vague or intuitive feeling. Accordingly, pure top-down and willingly controlled processing, such as free recall, should not have any access to sensory data. Empirical results from different research areas and from two experiments conducted by the authors are presented in this article to support these theoretical distinctions. The experiments were designed to separate a sensory-motor and a conceptual component in memory for two-digit numbers and two-letter items, when parts of the numbers or items were imaged or drawn on a tablet. The results of free recall and recognition are discussed in a theoretical framework which distinguishes sensory and conceptual information in memory.     

Saying and believing: the norm commonality assumption

Philosophical Studies - One very popular assumption in the epistemological literature is that belief and assertion are governed by one and the same epistemic norm. This paper challenges this claim....     

Pleasure and Aversion: Challenging the Conventional Dichotomy

Philosophy and its descendents in the behavioral sciences have traditionally divided incentives into those that are sought and those that are avoided. Positive incentives are held to be both attractive and memorable because of the direct effects of pleasure. Negative incentives are held to be unattractive but still memorable (the problem of pain) because they force unpleasant emotions on an individual by an unmotivated process, either a hardwired response (unconditioned response) or one substituted by association (conditioned response). Negative incentives are divided into those that are always avoided and those that are avoided only by higher mental processes—archetypically the passions, which are also thought of as hardwired or conditioned. Newer dichotomies within the negative have been proposed, hinging on whether a negative incentive is nevertheless sought (“wanted but not liked”) or on an incentive's being negative only because it is confining (the product of “rule worship”). The newer dichotomies have lacked motivational explanations, and there is reason to question conditioning in the motivational mechanism for the older ones.

Both experimental findings and the examination of common experience indicate that even the most aversive experiences, such as pain and panic, do not prevail in reflex fashion, but because of an urge to attend to them. The well-established hyperbolic curve in which prospective rewards are discounted implies a mechanism for such an urge, as well as for the “lower” incentives in the other dichotomies. The properties of these lower incentives are predicted by particular durations of temporary preferences on a continuum that stretches from fractions of a second to years.