A computational analysis of general intelligence tests for evaluating cognitive development

The progression in several cognitive tests for the same subjects at different ages provides valuable information about their cognitive development. One question that has caught recent interest is whether the same approach can be used to assess the cognitive development of artificial systems. In particular, can we assess whether the ‘fluid’ or ‘crystallised’ intelligence of an artificial cognitive system is changing during its cognitive development as a result of acquiring more concepts? In this paper, we address several IQ tests problems (odd-one-out problems, Raven’s Progressive Matrices and Thurstone’s letter series) with a general learning system that is not particularly designed on purpose to solve intelligence tests. The goal is to better understand the role of the basic cognitive operational constructs (such as identity, difference, order, counting, logic, etc.) that are needed to solve these intelligence test problems and serve as a proof-of-concept for evaluation in other developmental problems. From here, we gain some insights into the characteristics and usefulness of these tests and how careful we need to be when applying human test problems to assess the abilities and cognitive development of robots and other artificial cognitive systems.     

Small wonders of the Indian market

Heavy competition in India in almost all product categories, due to diversification by large and medium companies and increased entry of multinationals, has restricted the growth of domestic companies. Previously, large companies enjoyed high profit margins by targeting premium priced products in the upper strata of Indian society. High levels of competition from equally reputed brands have not only decreased the companies' market share but also created price wars, reducing profit margins and limiting market growth. This has motivated companies to consider the lower classes and the rural segments, which they had previously ignored. By targeting these segments with products in small packs at lower price points, companies have experienced great success. At the same time, small packs also pose some challenges for the companies. This paper explains the importance of small packs for market expansion in various product categories within the Indian market, drawing on several examples to support the views of the authors. Copyright © 2004 Henry Stewart Publications.     

Epistemology and artificial intelligence

In this essay we advance the view that analytical epistemology and artificial intelligence are complementary disciplines. Both fields study epistemic relations, but whereas artificial intelligence approaches this subject from the perspective of understanding formal and computational properties of frameworks purporting to model some epistemic relation or other, traditional epistemology approaches the subject from the perspective of understanding the properties of epistemic relations in terms of their conceptual properties. We argue that these two practices should not be conducted in isolation. We illustrate this point by discussing how to represent a class of inference forms found in standard inferential statistics. This class of inference forms is interesting because its members share two properties that are common to epistemic relations, namely defeasibility and paraconsistency. Our modeling of standard inferential statistical arguments exploits results from both logical artificial intelligence and analytical epistemology. We remark how our approach to this modeling problem may be generalized to an interdisciplinary approach to the study of epistemic relations.     

An evolutionary system for neural logic networks using genetic programming and indirect encoding

Nowadays, intelligent connectionist systems such as artificial neural networks have been proved very powerful in a wide area of applications. Consequently, the ability to interpret their structure was always a desirable feature for experts. In this field, the neural logic networks (NLN) by their definition are able to represent complex human logic and provide knowledge discovery. However, under contemporary methodologies, the training of these networks may often result in non-comprehensible or poorly designed structures. In this work, we propose an evolutionary system that uses current advances in genetic programming that overcome these drawbacks and produces neural logic networks that can be arbitrarily connected and are easily interpretable into expert rules. To accomplish this task, we guide the genetic programming process using a context-free grammar and we encode indirectly the neural logic networks into the genetic programming individuals. We test the proposed system in two problems of medical diagnosis. Our results are examined both in terms of the solution interpretability that can lead in knowledge discovery, and in terms of the achieved accuracy. We draw conclusions about the effectiveness of the system and we propose further research directions.     

Logic programs and connectionist networks

One facet of the question of integration of Logic and Connectionist Systems, and how these can complement each other, concerns the points of contact, in terms of semantics, between neural networks and logic programs. In this paper, we show that certain semantic operators for propositional logic programs can be computed by feedforward connectionist networks, and that the same semantic operators for first-order normal logic programs can be approximated by feedforward connectionist networks. Turning the networks into recurrent ones allows one also to approximate the models associated with the semantic operators. Our methods depend on a well-known theorem of Funahashi, and necessitate the study of when Funahashi's theorem can be applied, and also the study of what means of approximation are appropriate and significant.     

Methodological naturalism and epistemic internalism

Epistemic naturalism holds that the results or methodologies from the cognitive sciences are relevant to epistemology, and some have maintained that scientific methods are more compatible with externalist theories of justification than with internalist theories. But practically all discussions about naturalized epistemology are framed exclusively in terms of cognitive psychology, which is only one of the cognitive sciences. The question addressed in this essay is whether a commitment to naturalism really does favor externalism over internalism, and we offer reasons for thinking that naturalism in epistemology is compatible with both internalist and externalist conceptions of justification. We also argue that there are some distinctively internalist aims that are currently being studied scientifically and these notions, and others, should be studied by scientific methods. This essay is dedicated to Deborah Mayo, who has long advocated using error statistical techniques to analyze and resolve epistemological puzzles in the philosophy of science. This essay follows the same spirit by advocating that computational concepts and techniques be applied within the heart of traditional, analytic epistemology.     

Motor programming in apraxia of speech

Apraxia of Speech (AOS) is an impairment of motor programming. However, the exact nature of this deficit remains unclear. The present study examined motor programming in AOS in the context of a recent two-stage model [Klapp, S. T. (1995). Motor response programming during simple and choice reaction time: The role of practice. Journal of Experimental Psychology: Human Perception and Performance, 21, 1015–1027; Klapp, S. T. (2003). Reaction time analysis of two types of motor preparation for speech articulation: Action as a sequence of chunks. Journal of Motor Behavior, 35, 135–150] that proposes a preprogramming stage (INT) and a process that assigns serial order to multiple programs in a sequence (SEQ). The main hypothesis was that AOS involves a process-specific deficit in the INT (preprogramming) stage of processing, rather than in the on-line serial ordering (SEQ) and initiation of movement. In addition, we tested the hypothesis that AOS involves a central (i.e., modality-general) motor programming deficit. We used a reaction time paradigm that provides two dependent measures: study time (the amount of time for participants to ready a motor response; INT), and reaction time (time to initiate movement; SEQ). Two experiments were conducted to examine INT and SEQ in AOS: Experiment 1 involved finger movements, Experiment 2 involved speech movements analogous to the finger movements. Results showed longer preprogramming time for patients with AOS but normal sequencing and initiation times, relative to controls. Together, the findings are consistent with the hypothesis of a process-specific, but central (modality-independent) deficit in AOS; alternative explanations are also discussed.     

Change detection evokes a Simon-like effect

The present experiment studied choice response context effects on the programming of response sequences using behavioural and electrophysiological methods. Participants were asked to produce responses differing in sequence length (1-key vs. 3-key responses) with either their left or right hand in a choice reaction time (RT) task. The choice response context was manipulated by a blocked or mixed execution of 1-key and 3-key responses. A sequence length effect on RT was observed in the blocked but not in the mixed condition. The time course of the lateralized readiness potential indicates a motoric locus of the sequence length effect, suggesting that the response hand is activated before the entire motor program is established.     

LÉVY HIERARCHY IN WEAK SET THEORIES

We investigate the interactions of formula complexity in weak set theories with the axioms available there. In particular, we show that swapping bounded and unbounded quantification preserves formula complexity in presence of the axiom of foundation weakened to an arbitrary set base, while it does not if the axiom of foundation is further weakened to a proper class base. More attention is being paid to the necessary axioms employed in the positive results, than to the combinatorial strength of the positive results themselves.     

Propositions, Sets, and Worlds

If we agree with Michael Jubien that propositions do not exist, while accepting the existence of abstract sets in a realist mathematical ontology, then the combined effect of these ontological commitments has surprising implications for the metaphysics of modal logic, the ontology of logically possible worlds, and the controversy over modal realism versus actualism. Logically possible worlds as maximally consistent proposition sets exist if sets generally exist, but are equivalently expressed as maximally consistent conjunctions of the same propositions in corresponding sets. A conjunction of propositions, even if infinite in extent, is nevertheless itself a proposition. If sets and hence proposition sets exist but propositions do not exist, then whether or not modal realism is true depends on which of two apparently equivalent methods of identifying, representing, or characterizing logically possible worlds we choose to adopt. I consider a number of reactions to the problem, concluding that the best solution may be to reject the conventional model set theoretical concept of logically possible worlds as maximally consistent proposition sets, and distinguishing between the actual world alone as maximally consistent and interpreting all nonactual merely logically possible worlds as submaximal. I am grateful to the Netherlands Institute for Advanced Study in the Humanities and Social Sciences (NIAS), Royal Netherlands Academy of Arts and Sciences (KNAW), for supporting this among related research projects in philosophical logic and philosophy of mathematics during my Resident Research Fellowship in 2005-2006.