The previous post concluded that the Motivational Design Framework lacked a unifying architecture - that the synthesis of SDT, discrepancy theory, ARCS, flow, and Dreyfus was ad hoc rather than systematic, and that Ford (1992) had already demonstrated what a more rigorous integration might look like. This post picks up that thread, but takes it somewhere the earlier series couldn't: into the formal apparatus that the Planning and Design series has been building. If motivation is about the gap between where you are and where you want to be, and if conceptual spaces give us a geometry for representing where things are, then motivation should be expressible as a vector. That turns out to be a more productive thought than it first appears.
Ford's formula
Ford's (1992) Motivational Systems Theory, grounded in the Living Systems Framework, offers the clearest starting point. His definition of motivation has three components: personal goals (the direction), emotional arousal processes (the energisation), and personal agency beliefs (the confidence that action will produce results). Ford represents this multiplicatively - Motivation = Goals x Emotions x Personal Agency Beliefs - which means that if any component drops to zero, so does motivation. You can want something desperately and feel nothing will work; you can believe in your ability but have no goal worth pursuing; you can have both goal and confidence but be emotionally flat. The multiplication matters because it captures how motivation fails: not gradually but categorically, when any single component is absent.
Ford's unit of analysis is the "behaviour episode" - a context-specific, goal-directed pattern of behaviour that unfolds until the goal is achieved, attention is captured by something else, or the goal is judged unattainable. This is close to what Gärdenfors (2014) would call an event, and what Iqbal (2018) would frame as a service encounter. The convergence isn't accidental. All three are modelling the same thing: a bounded sequence of action directed toward a change in state.
Events as vectors
Gärdenfors's (2014) cognitive theory of events, developed through conceptual spaces, provides the geometric formalism that Ford's theory lacks. In Gärdenfors's account, a prototypical event involves an agent who generates a force vector that affects a patient, causing a change in their position within some quality space. The force has magnitude, direction, and a point of origin; the result is a movement - a result vector - representing the change of state the patient undergoes. Verbs, on this account, refer to either the force vector or the result vector, but not both.
This is directly relevant to motivation. Discrepancy theory - the engine of the MPhil's framework - says motivation arises from a gap between current state and desired state. In a conceptual space, that gap is a vector. It has direction (what kind of change is sought), magnitude (how far the desired state is from the current one), and a point of origin (where the actor currently sits in the relevant quality space). What the MPhil called "managing discrepancy" was, without quite knowing it, an attempt to calibrate the magnitude and direction of this vector - creating sufficient distance between where visitors were and where the exhibition wanted them to go, without creating so much that the gap became overwhelming.
The four kinds of event that Gärdenfors distinguishes - causation, self-propelled motion, action, and result - map onto different motivational structures. A caused event has its force vector originating outside the patient; a self-propelled event generates its own. Motivational design, on this reading, is the practice of arranging force vectors - some originating from the designed environment, some from other actors, some from the person's own goals - so that the resulting movement through the state space is one the person values. When the external force vectors align with the person's own self-propelled direction, that's what SDT calls autonomy support. When they override it, that's control.
Heckhausen's action phases
Heckhausen and Heckhausen (2010) add temporal structure to what the geometric account leaves static. Their Rubicon model distinguishes four phases of goal-directed action: predecisional (deliberating about which goal to pursue, weighing desirability against feasibility), preactional (planning how to achieve the chosen goal, waiting for the right moment), actional (executing the plan, managing effort and persistence), and postactional (evaluating what happened, adjusting for next time). The metaphor is Caesar's crossing of the Rubicon - the point at which deliberation becomes commitment, and motivation becomes volition.
The Rubicon matters here because it separates two things the MPhil conflated. Goal setting - the process by which someone comes to want something - is a motivational question. Goal striving - the process by which they pursue it - is a volitional one. The MPhil's framework was mostly about goal setting: creating conditions under which visitors would become interested, curious, engaged. It had much less to say about what happened after commitment - about how engagement persists, how effort is regulated, how people decide whether to continue or give up. In vector terms: the framework attended to the direction of the motivational vector (where are we pulling people?) but not to the dynamics of traversal (how does movement through the space actually unfold over time?).
Promises and motivation
Promise theory introduces something neither Ford nor Gärdenfors addresses: the relational structure between autonomous agents. In Burgess's (2017) formulation, a promise is a voluntary declaration by an autonomous agent about its own future behaviour. Promises can't be imposed; they can only be offered. This makes promise theory structurally aligned with SDT's autonomy requirement - both insist that the most productive forms of cooperation arise from voluntary commitment rather than external control.
Iqbal (2018) connects this to service design through four types of promise: demand for affordance, supply of affordance, demand for performance, and supply of performance. Crucially, Iqbal notes that "who and why describe the motivations of the side making a promise" while "how and what describe the expectations of the side accepting it." Each promise has a motivational dimension (why this agent commits) and an expectational dimension (what the other agent can count on). A service encounter, then, isn't a single force vector but an exchange of promises between agents, each with their own motivational vectors - their own goals, agency beliefs, and emotional investments.
This reframes the MPhil's core problem. The framework assumed a single motivational vector: the designer's intent to move the visitor from ignorance to engagement. But a service encounter involves at least two vectors - the provider's and the user's - and they needn't point in the same direction.
The exhibition wanted visitors to understand ergonomics as a process; visitors wanted to see interesting objects. The project team wanted to demonstrate public engagement for EPSRC reporting; the Design Museum wanted footfall and good press. These are different vectors in different quality spaces, and the "motivational design" challenge isn't to define one vector and optimise for it but to negotiate alignment among several.
Ford (1992) has a concept for this: goal alignment versus goal conflict. "Achievement and competence are greatly facilitated by efforts to enhance goal alignment, both in terms of the goals afforded by the context and the goal coordination skills of the individual." The attention economy is a case of structural goal conflict - platform goals (attention extraction) systematically oppose user goals (time well spent). The MPhil's exhibition was a milder case: institutional goals partially aligned with visitor goals but also diverged from them in ways the framework couldn't represent because it only modelled one agent's motivation at a time.
What a motivational vector would look like
If we take the geometric framing seriously, a motivational vector for an actor in a service encounter would need to represent at least three things: the actor's current position in a relevant state space (what they know, what they can do, how they feel about it), the position they're oriented toward (their goal, whether explicitly formulated or tacitly held), and the force available to close the gap (their confidence, emotional engagement, and the affordances the environment provides). Ford's multiplicative formula gives the scalar magnitude - how motivated the person is. The conceptual space gives the geometry - in what direction and through what dimensions.
For the ergonomics exhibition, this might have looked something like: visitors arrive with a position in an "understanding of ergonomics" quality space clustered around chairs and comfort (the empirical finding). The exhibition attempts to generate a result vector that moves them toward a region characterised by process, system, and human-centred design. The force vector comes partly from the exhibition design (external) and partly from visitors' own curiosity and prior interests (internal, self-propelled).
The challenge Heckhausen would identify is that most visitors were in the predecisional phase - they hadn't committed to the goal of understanding ergonomics differently; they were browsing. The exhibition needed to do motivational work (making the goal desirable) and volitional work (sustaining engagement once initial curiosity was piqued), and the framework didn't distinguish between these.
The promise-theoretic dimension adds the relational layer. The exhibition implicitly promised visitors an engaging experience in exchange for their attention and openness. Visitors implicitly promised to engage with exhibits in exchange for intellectual stimulation or entertainment. When these promises aligned (exhibits that were both scientifically substantive and experientially rewarding), the service encounter worked. When they diverged (exhibits that were scientifically important but experientially flat), the motivational vectors pointed in different directions and engagement stalled.
Was this the MPhil's project all along?
In a sense, yes. The MPhil was trying to model the motivational state of actors in designed experiences - to represent where they were and where they might go, and to understand the forces that could move them. It lacked the formal apparatus to do this: no geometric representation of states, no vector model of forces, no relational structure for multiple agents, no temporal model for how goal pursuit unfolds. It had the psychological content - SDT, discrepancy, flow, ARCS, Dreyfus - but not the architecture to hold it together.
The Planning and Design series has been building that architecture from a different direction. Conceptual spaces provide the geometry. State spaces provide the landscape. Promise theory provides the relational structure. Gärdenfors's event semantics provide the dynamics. What this post suggests is that these formal tools might retrospectively supply what the MPhil needed and didn't have: a way to represent motivation not as a checklist of psychological factors but as directed movement through a structured space, generated by forces that can be analysed, compared, and - in the service design case - negotiated between agents.
Whether this amounts to anything more than a pleasing analogy is an open question. The danger is exactly what the previous post warned about: overestimating the utility of frameworks, mistaking a way of seeing for a way of doing. A vector model of motivation is still a model. It might sharpen the questions - whose vector? in what space? with what force? - without producing better answers. But it connects the psychological substance of the MPhil to the formal apparatus of the wider project in a way that feels less like retrospective tidying and more like recognising what was there all along, waiting for a language adequate to express it.
References
Burgess, M. (2017). A Treatise on Systems. Self-published.
Ford, M.E. (1992). Motivating Humans: Goals, Emotions, and Personal Agency Beliefs. Sage.
Gärdenfors, P. (2014). The Geometry of Meaning: Semantics Based on Conceptual Spaces. MIT Press.
Heckhausen, J. and Heckhausen, H. (2010). Motivation and Action (2nd ed.). Cambridge University Press.
Iqbal, M. (2018). Thinking in Services: Encoding and Expressing Strategy Through Design. BIS Publishers.