Decision field theory

From Wikipedia, the free encyclopedia

Decision Field Theory (DFT), is a dynamic - cognitive approach to human decision making. It is a cognitive model that describes how people make decisions rather than a rational model that prescribes what people should do. It is also a dynamic model of decision making rather than a static model, because it describes how a person's preferences evolve across time until a decision is reached rather than assuming a fixed state of preference. The preference evolution process is mathematically represented as a stochastic process called a diffusion process. It is used to predict how humans make decisions under uncertainty, how decisions change under time pressure, and how choice context changes preferences. This model can be used to predict not only the choices that are made but also decision or response times. The Decision Field Theory (DFT) was published by Jerome R. Busemeyer and James T. Townsend in 1993.[1] The DFT has been shown to account for many puzzling findings regarding human choice behavior including violations of stochastic dominance, violations of strong stochastic transitivity, violations of independence between alternatives, serial position effects on preference, speed accuracy tradeoff effects, inverse relation between probability and decision time, changes in decisions under time pressure, as well as preference reversals between choices and prices. The DFT also offers a bridge to neuroscience. Recently, the authors of decision field theory also have begun exploring a new theoretical direction called Quantum Cognition.

   

1 Decision Field Theory [2] 2 Decision Field Theory - Explaining Context Effects 3 DFT and Neuroscience [14] 4 References


Field Theory [2]

The name of decision field theory was chosen to reflect the fact that the inspiration for this theory comes from an earlier approach - avoidance conflict model contained in Kurt Lewin's general psychological theory, which he called field theory. DFT is a member of a general class of sequential sampling models that are commonly used in a variety of fields in cognition [3] [4] [5] [6] [7] [8] .[9]The basic ideas underlying the decision process for sequential sampling models is illustrated in Figure 1 below. Suppose the decision maker is initially presented with a choice between three risky prospects, A, B, C, at time t = 0. The horizontal axis on the figure represents deliberation time (in seconds), and the vertical axis represents preference strength. Each trajectory in the figure represents the preference state for one of the risky prospects at each moment in time.

prolonging the deliberation process. at each moment in time. Low thresholds allow a weak preference state to determine the decision. In this example. and increasing accuracy. the decision maker thinks about various payoffs of each prospect. Choice probability is determined by the first option to win the race and cross the upper threshold. Thus each action is defined by a probability distribution across these four outcomes. and also suppose for simplicity that there are only four possible final outcomes. but later (after 600 ms) attention is shifted toward advantages favoring prospect A. which produces an affective reaction. and decision time is equal to the deliberation time required by one of the prospects to reach this threshold. to each prospect. which in this case is prospect A after about one second.[10] High thresholds require a strong preference state to be reached. Under high time pressure. and impulsive and careless decision makers use a low threshold. and decreasing accuracy. The threshold is an important parameter for controlling speed–accuracy tradeoffs. Very careful and deliberative decision makers tend to use a high threshold. These valences are integrated across time to produce the preference state at each moment. during the early stages of processing (between 200 and 300 ms). then prospect B would be chosen instead of prospect A (and done so earlier). The affective values produced by .50) in Figure 1. a higher threshold can be used to increase accuracy. assume that the decision maker has a choice among three actions. If the threshold is set to a lower value (about .0 in this example): the first prospect to reach the top threshold is accepted. decision makers must choose a low threshold. which allows more information about the prospects to be sampled. attention is focused on advantages favoring prospect B. shortening the deliberation process. or valence. The stopping rule for this process is controlled by a threshold (which is set equal to 1. but under low time pressure.Intuitively. To provide a bit more formal description of the theory. which cuts off sampling information about the prospects. Thus decisions can reverse under time pressure.

In other words. However. Values of sii < 1 suggest decay in the memory or impact of previous valences over time. this is a Markov process. if the probability of choosing option X is greater than option Y when only X. and a new option Z. The output preference state for action i at time t is symbolized as Pi(t). which produces a momentary evaluation. empirical findings obtained by consumer researchers . adding an option should not change the preference relation between the original pair of options.(t) equals the average across all the momentary actions. for action i. adding an option should only decrease the probability of choosing one of the original pair of options. Finally. whereas values of sii > 1 suggest growth in impact over time (primacy effects).(t).Y are available. These lateral inhibitory coefficients are important for explaining context effects on preference described later. The momentary evaluation of each action is compared with other actions to form a valence for each action at each moment. The dynamic system is described by the following linear stochastic difference equation for a small time step h in the deliberation process: Pi(t+h) = Σ sijPj(t)+vi(t+h). and according to this principle. At any moment in time. where U. The decision field theory can also be seen as a dynamic and stochastic random walk theory of decision making.Explaining Context Effects The DFT is capable of explaining context effects that many decision making theories are unable to explain. then option X should remain more likely to be chosen over Y even when a new option Z is added to the choice set. vi(t) = Ui(t) – U. The total valence balances out to zero so that all the options cannot become attractive simultaneously. Ui(t). Y. for payoff j offered by action i. is assumed to fluctuate according to a stationary stochastic process. The magnitudes of the lateral inhibitory coefficients are assumed to be an increasing function of the similarity between choice options. matrix formulas have been mathematically derived for computing the choice probabilities and distribution of choice response times. the probability of choosing option X from a set containing only X and Y should be greater than or equal to the probability of choosing option X from a larger set containing options X. Formally. A second principle is called regularity. causing changes in the anticipated payoff of each action across time. the decision maker anticipates the payoff of each action. controls the memory for past input valences for a preference state. One principle is called independence of irrelevant alternatives.each payoff are represented by the values mj. This momentary evaluation is an attention-weighted average of the affective evaluation of each payoff: Ui(t) = Σ W ij(t)mj. the valences are the inputs to a dynamic system that integrates the valences over time to generate the output preference states. produce competition among actions so that the strong inhibit the weak.The positive self feedback coefficient. In other words. The negative lateral feedback coefficients. Many classic probabilistic models of choice satisfy two rational types of choice principles. The valence represents the momentary advantage or disadvantage of each action. In other words. This reflects the idea that attention is shifting from moment to moment. W ij(t). sij = sji < 0 for i not equal to j. then this moderates the preference for other actions. [edit]Decision Field Theory . as preference for one action grows stronger. and according to this principle. presented as a model positioned between lower-level neural activation patterns and more complex notions of decision making found in psychology and economics. sii = s > 0. The attention weight at time t.

but when S is added. This is another violation of the independence from irrelevant alternatives property because X is chosen more often than C in a binary choice. but costs more than the BMW. This effect occurs when an option C is added that is a compromise between X and Y. X is chosen more frequently than Y. The Audi is similar to the BMW because both are not very economic but they are both high quality and sporty. The Ford focus is different from the BMW and Audi because it is more economical but lower quality. The second context effect is the compromise effect. and this attraction effect violates the principle of regularity. Next suppose a new choice set is formed by adding an option S that is similar to X. This violates the independence from irrelevant alternatives property because in a binary choice. suppose X is a BMW.D is larger than the probability of choosing X from a set containing only X and Y.Y. and S is an Audi. X (BMW) is chosen more often than C (Honda). and if the lateral connections are all set to zero. then C is chosen more often than X. The first context effect is the similarity effect. If the attention switching process is eliminated. then option C (Honda) becomes the compromise between X (BMW) and Y (Ford Focus). there is little or no reason to choose D over X. then Y is chosen more frequently than X. then C = Honda becomes a compromise between X = BMW and Y = Ford Focus. and in this situation D is rarely ever chosen over X. and C is then chosen more frequently than X. Suppose in a binary choice. adding D to a choice set boosts the probability of choosing X. If X is similar to S. This effect occurs when the third option D is very similar to X but D is defective compared to X. However. According to DFT. This causes the probability of choosing X to drop below Y when S is added to the choice set. The contrast effects produced . This effect occurs with the introduction of a third option S that is similar to X but it is not dominated by X. DFT accounts for all three effects using the same principles and same parameters across all three findings. then the attraction and compromise effects disappear. For example D may be a new sporty car developed by a new manufacturer that is similar to option X = BMW. Suppose in a binary choice. the probability of choosing X from a set containing X. and both are very different from Y. then the similarity effect disappears. Thus the probability of Y remains the same whether S is presented as an option or not. the attention switching mechanism is crucial for producing the similarity effect. when choosing between C = Honda and X = BMW. the people tend to view X and S as one group and Y as another option. However. The defective option D makes X shine. X is chosen more frequently than Y. if another option Y = Ford Focus is added to the choice set.studying human choice behavior have found systematic context effects that systematically violate both of these principles. But when option Y (Ford Focus) is added to the choice set. Therefore. but C when option Y is added to the choice set. However. the latter is less economical but higher quality. For example. The third effect is called the attraction effect. For example. In particular. This property of the theory entails an interesting prediction about the effects of time pressure on preferences. which says that adding another option cannot increase the popularity of an option over the original subset. the probability of X will decrease by approximately half with the introduction of S. Y is a Ford focus. but the lateral inhibitory connections are critical for explaining the compromise and attraction effects.

[9] and categorization. diffusion models provide the potential to form a theoretical bridge between neural models of sensory-motor tasks and behavioral models of complex-cognitive tasks. In addition to these neuroscience applications.[5][6][8] to memory recognition.[3][4] Thus. Empirical tests show that prolonging the decision process increases the effects[12] and time pressure decreases the effects. except that the dynamics are approximated by linear systems. A conclusion that one can draw is that the neural areas responsible for planning or carrying out certain actions are also responsible for deciding the action to carry lateral inhibition require time to build up. analogues) have been used by cognitive scientists to model performance in a variety of tasks ranging from sensory detection. Typical findings indicate that neural activation regarding stimulus movement information is accumulated across time up to a threshold. such as the decision field theory. The decision processes of sensory-motor decisions are beginning to be fairly well understood both at the behavioral and neural levels. diffusion models (or their discrete time. random walk. as well as the choice and response time distributions.[11]). Alternatively.[7] and perceptual discrimination. and a behavioral response is made as soon as the activation in the recorded area exceeds the threshold (see [15] [16][17] [18] [19] for examples). can be viewed as stochastic recurrent neural network models. Diffusion models. a decidedly embodied notion. Recent studies that record neural activations in non-human primates during perceptual decision making tasks have revealed that neural firing rates closely mimic the accumulation of preference theorized by behaviorally-derived diffusion models of decision making. The linear approximation is important for maintaining a mathematically tractable analysis of systems perturbed by noisy inputs. the spike activation pattern. Mathematically.especially in Two-alternative forced choice tasks(see Smith & Ratcliff[20] for a summary). then these effects should get larger under time pressure. if context effects are produced by switching from a weighted average rule under binary choice to a quick heuristic strategy for the triadic choice.[13] [edit]DFT and Neuroscience [14] The Decision Field Theory has demonstrated an ability to account for a wide range of findings from behavioral decision making for which the purely algebraic and deterministic models often used in economics and psychology cannot account. . can be well described by what are known as diffusion models . which implies that the attraction and compromise effects should become larger under prolonged deliberation (see Roe et al.