 |
| David Cuykendall Click for Index Page |
The term ‘heuristic’ is of Greek origin, meaning ‘‘serving to find out or discover.’’ Heuristic methods need to be distinguished from optimizing methods. Optimal decision making depends on finding answers through analyzing massive quantities of data, the handling of which is computationally expensive. In contrast, heuristic methods seek ‘good enough’ answers and depend on simple rules.
Most problems of interest are computationally intractable, and this is why engineers and artificial intelligence researchers often rely on heuristics to make computers smart.
The development of an understanding of why and when heuristics are often more accurate than strategies that use more information and computation rests not in the heuristics alone, but in the match between the heuristics and the environment that underlies the problem space.
Simple heuristics are perhaps best understood from the perspective of pattern recognition, where there are many examples of how biased inferences can predict more accurately than unbiased deductions, provided the ‘bias’ is based on a pattern that is ‘ecologically valid’; that is — if the pattern the heuristics rely upon accurately matches the environment from which the pattern is drawn.
However, if the pattern in the environment is unknown — meaning it remains to be discovered —computationally expensive optimizing methods are necessary.
Heuristics rely on ‘good enough’ explanations that could be in error upon the discovery of additional information; they rest upon ‘justification’ and not ‘explanations’ that can be logically proved.
A further difference between heuristic and optimizing methods rest in the standards that are applied to them. Heuristic methods rely on inferences (inductions) the supply reasons for their suitability. They don’t guarantee the claims they make, whereas optimizing methods rest on deductions that are valid or not. Heuristic methods have varying degrees of strength. Optimizing methods either guarantee the most optimal answer or not.
Satisficing is a decision-making strategy or cognitive heuristic that entails searching through the available alternatives until an acceptability threshold is met. This is contrasted with optimal decision making, an approach that specifically attempts to find the best alternative available.
The term satisficing, a portmanteau of satisfy and suffice, was introduced to explain the behavior of decision makers under circumstances in which an optimal solution cannot be determined. We can rarely evaluate all outcomes with sufficient precision, usually do not know the relevant probabilities of outcomes, and possess only limited memory.
In decision making, satisficing explains the tendency to select the first option that meets a given need or select the option that seems to address most needs rather than the "optimal" solution.
Example: A task is to sew a patch onto a pair of jeans. The best needle to do the threading is a 4 inch long needle with a 3 millimeter eye. This needle is hidden in a haystack along with 1000 other needles varying in size from 1 inch to 6 inches. Satisficing claims that the first needle that can sew on the patch is the one that should be used. Spending time searching for that one specific needle in the haystack is a waste of energy and resources.
Satisficing also occurs in consensus building when the group looks towards a solution everyone can agree on even if it may not be the best.
Example: A group spends hours projecting the next fiscal year's budget. After hours of debating they eventually reach a consensus, only to have one person speak up and ask if the projections are correct. When the group becomes upset at the question, it is not because this person is wrong to ask, but rather because the group has already come up with a solution that works. The projection may not be what will actually come, but the majority agrees on one number and thus the projection is good enough to close the book on the budget.
In many circumstances, the individual may be uncertain about what constitutes a satisfactory outcome.
Example: An individual who only seeks a satisfactory retirement income may not know what level of wealth is required—given uncertainty about future prices—to ensure a satisfactory income. In this case, the individual can only evaluate outcomes on the basis of their probability of being satisfactory. If the individual chooses that outcome which has the maximum chance of being satisfactory, then this individual's behavior is theoretically indistinguishable from that of an optimizing individual under certain conditions.
Satisficing is often a good option when making a decision, but it can also be detrimental if used the wrong way.
Example: When considering a medical issue such as a diagnosis, satisficing is not the best decision making strategy to use. On the other hand, when choosing an outfit or an option from a menu, it can be helpful. When there is an unlimited amount of information available and it is necessary to eliminate options, satisficing is beneficial because it helps the person making the decision effectively and efficiently reach a conclusion.
Satisficing and optimization
One definition of satisficing is that it is optimization where all costs, including the cost of the optimization calculations themselves and the cost of getting information for use in those calculations, are considered. As a result, the eventual choice is usually sub-optimal in regard to the main goal of the optimization, i.e., different from the optimum in the case that the costs of choosing are not taken into account.
Satisficing as a form of optimization
Alternatively, satisficing can be considered to be just constraint satisfaction, the process of finding a solution satisfying a set of constraints, without concern for finding an optimum.
In economics, satisficing is a behavior which attempts to achieve at least some minimum level of a particular variable, but which does not necessarily maximize its value. The most common application of the concept in economics is in the theory that postulates producers treat profit not as a goal to be maximized, but as a constraint. Under this theory, a critical level of profit must be achieved; thereafter, priority is attached to the attainment of other goals.
The aspiration level is the payoff aspired to. What determines the aspiration level? This can come from past experience (some function of previous payoffs), or some organizational or market mandate.
Conclusion
The distinction between satisficing and maximizing not only differs in the decision-making process, but also in the post-decision evaluation. Maximizers (optimizers) tend to use a more exhaustive approach to their decision-making process: they seek and evaluate more options than satisficers do to achieve greater satisfaction. However, whereas satisficers tend to be relatively pleased with their decisions, maximizers tend to be less happy with their decision outcomes. This is thought to be due to limited cognitive resources people have when their options are vast, forcing maximizers to not make an optimal choice.
