One of the most prolific examples of a failure of deductive reasoning can be seen in the concept of the social contract algorithm. The social contract algorithm is designed to frame questions within a given context and illicit responses from individuals based on the notion that those individuals will respond to the questions based on what an individual is entitled to do or feels obligated to do (Parsell, 2005). This algorithm is utilized most often to detect cheating behavior in that it questions the notion of entitlement and predicates that cheaters are individuals who will reap the rewards of a social contract without fulfilling the obligation on which the contract is based.
Sugiyama, Tooby & Cosmides (2002) examined the notion that the social contract algorithm was one with cross-cultural implications. Essentially, they posit that the ability to detect whether an individual is deceitful is one that is an inextricable component of human nature and describes humans as beings who are naturally “programmed” to be lie detectors. The process by which this is facilitated is one that utilizes deductive reasoning whereby certain facts about the society are accepted, and others are construed based on the accepted facts.
When one examines the inherent problems with the social contract algorithm and its application of deductive reasoning, one can clearly see that there are problems that can arise as a direct result of errors in comprehension, heuristic inadequacies, and errors in processing. The errors in comprehension refer to the notion that when individuals are faced with conceptualizing constructs, they first establish an understanding of what is meant by the constructs.
When there is a misunderstanding of the tasks involved in the process of reasoning, one can only get a flawed deductive reasoning process. Another intrinsic problem with deductive reasoning is one that involves the heuristics of deductive reasoning. In examining heuristics, deductive reasoning utilizes many of the constructs of the mental model approach. The mental model approach posits that errors in reasoning are made when an individual lacks a full set of mental models to represent all the components of a construct (Cooke, Gorman & Duran, 2007).
In order to improve the outcome of deductive reasoning applications, it is prudent that there is the remove a great deal of the ambiguity present in the constructs. This ambiguity contributes to the lack of comprehension and can be removed by establishing constructs that are clear and are limited in the various ways in which the constructs can be interpreted. Another method of improving deductive reasoning applications is to assure that there is an abundance of mental models which can paint a clearer picture of the constructs in question.
In examining inductive reasoning, it is prudent that we say that by its very nature, deductive reasoning poses several problems. Deductive reasoning operates to generalize conclusions based on specific cases. In examining this notion, one can clearly see that inherent in this process, there are some built-in inaccuracies in that specific cases may not be generalizable to the population at large. Gendler (2007) examined the different forms of reasoning and noted that inductive reasoning often employs associative techniques in order to arrive at its generalization. In so doing, the principles of congruity and similarity are applied, and an individual may resort to personal experience as an information source when attempting to generalize.
Additionally, there is a great reliance on generic concepts, images stereotypes. Examining this construct in terms of heuristics, one can clearly see the flaw in the process in that we are beginning with imprecise constructs, and the only thing we can extract from this is imprecise results. In addition to this, there are inherent problems with regards to bias as a great deal of the generalization in typical reasoning has a base in an individual’s personal experiences. One remedy for the problems posed by inductive approaches can be seen in an increase in the number of cases on which the generalizations are based.
The rationale behind this is the notion that with a larger body of empirical research comes more knowledge about the research subject. This increase in knowledge implies more precise generalizations. Another method for improving the results obtained from inductive approaches is one that attempts to reduce some of the biases inherent by impacting measures that facilitate a researcher’s awareness of his/her bias. This can be done utilizing similar approaches as with multicultural psychology.
After having examined both deductive and inductive approaches to assessing cognitive functioning, one can clearly see that both approaches are problematic to some degree. Both approaches exhibit heuristic problems; however, deductive approaches exhibit errors incomprehension.
Inductive approaches, on the other hand, exhibits researcher bias as one of its most fundamental problems. When choosing an approach, it is prudent that one takes these problem areas into consideration and adopt a methodology that minimizes the possibilities of problems. This decision is unique in that it has to take into consideration the nature of the investigation and determine the appropriate methodology based on an intricate combination of the research problem, variables, and the likelihood of experiencing errors based on the approach chosen.
References
Cooke, N.J., Gorman, J.C. & Duran, J.L. (2007). Team cognition in experienced command-and-control teams. Journal of Experimental Psychology: Applied, 13(3), 146-157.
Gandler, T.S. (2007). Philosophical Thought Experiments, Intuitions, and Cognitive Equilibrium. Midwest Studies In Philosophy, 31(1), 68-89.
Parsell, M. (2005). Content-sensitive Inference, Modularity and the Assumption of Formal Processing. Philosophical Psychology. 18(1), 45–58.
Sugiyama, L.S. , Tooby, J. & Cosmides, L. (2002). Cross-cultural evidence of cognitive adaptations for social exchange among the Shiwiar of Ecuadorian Amazonia. Proceedings of the National Academy of Sciences, 99(17), 11537–11542.