Heuristic Decision Process: Taking it Apart

     Throughout life, we are faced with particular problems that require a certain course of action. These problems or events that require a decision on our behalf must be dealt with accordingly if we are to make the correct decision. Mathematical problems require a certain algorithm that will always bring out a solution. However, these situation differ from scenarios in life outside of mathematics. When one is faced with a scenario that is familiar to past events, a decision can be made empirically by relying on the action that was taken before (if it led to a favorable outcome). The rule of thumb (heuristic) that was created by the past event's outcome is typically selected by the individual because it creates a solution faster than any other method. Most heuristics are simple everyday occurrences; one rule of thumb for someone trying to lose weight might be to ensure that their caloric intake isn't higher than the amount of calories they use in a day. 

     However, at times we are faced with situations that might seem familiar, but are truly novel by nature. In these situations, our previous pattern of providing a certain response to a particular condition does not work. If a problem proves to be novel, we must turn to an alternative decision making process.

  Hill-Climbing Heuristic

The hill-climbing heuristic is similar to the method of trial and error. Using the hill-climbing method, a person generally picks what appears to be the most direct route to the goal at each step. If this choice proves to be incorrect, the person might choose an alternative method to see if it achieves the goal faster. The main disadvantage to this method is that one is only able to see the goal and one step ahead of where they are at that moment. This method is appropriate only for certain tasks though. One example of a type of problem that requires the hill-climbing method is a maze.

Maze A                                                               Maze B

Each maze is unique, one method to solve it is through the hill-climbing method to get closer to the end of the maze.

  Means-Ends Heuristic

     Another problem solving method used is the means-ends heuristic. This method of problem solving involves identifying the main goal of the problem and then dividing that goal into sub-goals to achieve the end. Each sub-goal is a step to the ultimate goal of the problem; because of this, each step is necessary to achieve the final goal. An example of the means-ends heuristic is the game "Towers of Hanoi" which involves discs. The object of the game is to move all the discs from one peg to another without placing a larger disc on a smaller disc. See the video below for a demonstration of the process involved to solve the puzzle.

  Analogy Approach

     The most commonly used heuristic is the analogy approach. This method to problem solving involves using an old solution to solve a new problem and find a new solution. In other words, a previous problem is solved effectively and the content learned from the old problem is incorporated into the new problem to find a new solution. In order to use an old solution for a new problem, one must be able to distinguish the basic foundation of the problem at hand and draw a correlation from the new problem to the old problem. This basic foundation is referred to as a "problem isomorph". An example of this can be seen by Joe, an experienced cook who has learned from his past. After cooking his first meal that used cooking oil, Joe put the skillet into the sink and turned on the water before letting the dish cool down. The oil reacted vigorously with the water and scared Joe tremendously. When Joe prepared his next meal that used cooking oil, he let the skillet cool down before putting it into the sink. The use of his previous knowledge provides a solution to an analogous problem at hand.

Links:

MIND Reviews: On Second Thought : On the nature of heuristics and identifying that we do use "mind" shortcuts. It stipulates that heuristics can sometimes lead to quick reasoning that can be bad, and useful at times when the process is automatic and thus frees the mind to use its energy elsewhere.

GPS: General Problem Solver : This is one of the first AI programs that simulates human problem solving using the means-ends heuristic method. This program gave way to more sophisticated programs such as SOAR.

Require Students to use Problem Solving Steps : This encourages the teaching of problem solving strategies to students. Specifically, the analogy approach for teachers is encouraged so they are able to guide students in the right direction from their own experience.

Problem Solving in Psychology : A well rounded site that provides a wonderful ground work on the types of problem solving that is available. Provides excellent examples and has nice images to go along with it.

 ***To my knowledge, all of the images I have used are not under copyright. Please contact me at 10522708@live.mercer.edu if any of these images are being used inappropriately***