Slade, Stephen - The Interpersonal Model of Goal-Based Decision Making

The Structure of Concern Project compares many theoretical models from many disciplines to the Adizes PAEI model, arguing that they must all be reflecting the same underlying phenomenon. One concern structure model is described below.

The Interpersonal Model of Goal-Based Decision Making is part of a research program in artificial intelligence undertaken by Stephen Slade (Slade, 1992). It has been implemented in a specific AI program called VOTE, which models US Congressional roll call voting. Slade rejects standard prescriptive decision making models based on probabilities of outcomes and payoffs, focusing instead on pragmatic decision constraints such as resource limitations across a multitude of goals, and a resultant need for various kinds of planning.

Many models of personal decision making only touch upon three areas of the structure of concern: P, A and E – the three more individualistic styles. Slade’s model is notable for its inclusion of I. He asserts that an agent’s goals “include both personal goals and adopted goals derived from interpersonal relationships” (Slade, 1992). Goals and relationships form the two highest levels of his conceptual framework for this project, as its title would suggest.

Goals do not stand alone in this model. They are embedded in a ‘goal triad’ along with plans and resources. P, A and E can be roughly associated with Plans, Resources and Goals, respectively, but a closer analysis provides a more exact mapping.

P – Bottom-up Goal Development: This is an opportunistic decision making mode. When an agent detects or suspects the presence of a resource, he or she may then adopt a plan to acquire that resource. This resource-based planning may not be instrumental to any currently active higher-level goal. This style of planning is a more reliable source of repeated gratification and success, because the resource is available from the beginning. Since planning is the active moment of opportunistic decision making, P can be roughly associated with the planning point of the goal triad – where goals and resources are already largely given by context.

A – Resources: Resources give rise to goals and plans, but the converse is also true, goals and plans require resourcing decisions to be made, to maximize the return upon their investment. This kind of economic reasoning falls into the A domain. Resources include time, money, attention, skills, commitment, locations, space, relations, health, objects, information, natural resources and social power. They differ by being variously perishable, expendable, critical, fungible, costly, accessible, renewable, interleaving (simultaneously usable in more than one plan, e.g. location), proprietary, transferable, and inherently associated with certain plans and contexts. Resources pose an allocation problem, forcing the prioritization and ranking of goals and plans, and vigilance against waste.

E – Top-down Goal Development: High-level goals are established by the agent, determining a hierarchy of sub-goals for reaching the high-level goals. This kind of goal-based planning has no necessary relationship with success or gratifying outcomes. An agent may very well hold highly-valued goals with none of the resources or preconditions necessary for achieving it. The agent must acquire these resources, and there is always a chance of failure. This kind of goal-dependent, context-independent planning falls within the E domain.

I – Relationships: Goal-based decision making is strongly affected by interpersonal interaction. We recruit others towards achieving our own goals, and they do the same in return. Some relationships cause us to develop new personal goals. The importance of a relationship influences our prioritization of each others’ goals and the resources we are willing to allocate to attainment plans. There are also generalized expectations of reciprocity, and perceptions of symmetry or asymmetry that influence goal, resource and plan involvements with others. Slade suggests that adopted goals “are processed in the same way as personal goals, but their priority is determined by the importance and context of the relationship” (Slade, 1992).

1. Slade, Stephen. (1992). Goal-based Decision Making: An Interpersonal Model. (Report No. 892). New Haven, Connecticut: Yale University Department of Computer Science.
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License