Associations: why people want, avoid or choose something

Theories of semantic memory use the idea of spreading activation to explain how one concept can make related knowledge more accessible. Collins and Loftus (1975) proposed an influential account of semantic processing. As a useful illustration, the word 'holiday' might bring to mind rest, freedom, family, expense or crowds. The particular connections depend on experience and the immediate task.

Attitude research offers another distinction. Fazio (2007) describes attitudes as object-evaluation associations that differ in strength. Some evaluations can be activated relatively quickly when a person encounters an object; others require more deliberate assessment. Gawronski and Bodenhausen (2006) explain why associative reactions and consciously expressed judgments do not necessarily change together.

None of this implies that associations determine every decision. People can question a first impression, compare evidence, consider social expectations or choose not to act. Nor is an association network a map of physical neural connections. It is a cognitive or analytical model used to understand relationships among meanings.

For Neurofactor, the difference between meaning, evaluation, decision and behaviour is central. Treating all four as interchangeable makes almost any outcome seem explainable, but leaves little room for a genuinely testable intervention.

Associations can be shaped by repeated experience, language, stories, social interaction, previous outcomes and the setting in which something is encountered. A brand that consistently delivers reliably may develop an association with trust. A poorly handled interview might change someone's expectations of an employer. Knowledge, company and atmosphere can shape a museum visit in ways that differ from a previous visit.

Repetition does not guarantee a stronger positive association. A campaign can become tiresome, reinforce doubts or link the brand with an unintended meaning. People can also explicitly re-evaluate new information. Associative and reflective processes can reinforce or contradict one another (Gawronski & Bodenhausen, 2006).

The useful question is therefore not, 'How do we implant an association in the brain?' That suggests a level of precision and control that research does not support. A better question is: what meaning is already present, when does it become accessible, what experience might change it and how would we test whether that change occurred?

This series uses five connected questions. They are editorial and research lenses, not mutually exclusive categories of brain activity. A single decision may involve several of them.

Research question	What we want to understand	Possible research approach	Common overinterpretation
Attraction	What makes this worth choosing for this audience?	Explore desirable meanings, motives, evaluations and context	Confusing liking with actual choice
Resistance	Which expectation, concern or friction holds someone back?	Examine ambivalence, perceived risks and practical barriers	Assuming every non-action is rooted in fear
Differentiation	Why does this offer mean something different from alternatives?	Compare relevant and distinctive brand or offer associations	Mistaking uniqueness for preference
Persuasion	What evidence or experience makes a claim credible?	Test expectations, evidence needs, congruence and interpretation	Mistaking attention or understanding for belief
Behaviour change	What must change for the desired action to happen?	Combine intention, associations, habits, capability and environment	Treating a new attitude as changed behaviour

Attraction and resistance may grow from the same need. The desire to belong may make playing music together appealing while increasing concern about not fitting in. That is why positive and negative meanings should not always be compressed into one simple scale.

Differentiation is more than standing out. Being associated with something nobody values provides little competitive advantage. Keller's (1993) influential account of customer-based brand equity highlights associations that are strong, favourable and unique. The meaning of the difference matters more than merely having a difference.

Persuasion needs relevant evidence. A statement such as 'we are safe' may inspire confidence, but it may also raise questions: why are they emphasising safety so much? A proof point only becomes persuasive when it addresses the uncertainty a person is trying to resolve.

Behaviour change requires more than a better story. A favourable attitude or stated intention does not guarantee action if opportunity, skills or routines are missing. Research on the intention-behaviour gap and habits supports this distinction (Ajzen, 1991; Sheeran & Webb, 2016; Wood & Rünger, 2016).

An associative question becomes more useful when the choice situation is specific. 'The target audience' does not exist outside a context. What someone associates with a brand while seeing an advert may play a different role during a purchase, job application or physical experience.

Field	Recognisable tension	Association question	Behavioural outcome worth testing
Marketing	People know a brand but do not prefer it	Is the intended brand meaning relevant and distinctive?	Brand preference and choice among realistic alternatives
Sales	A customer sees the value but postpones a decision	What risk expectation or organisational obstacle is involved?	Decisions, evidence needs and conversion in the actual buying journey
Recruitment	Candidates like the employer but do not apply	What does the job, culture and application process mean to them?	Suitable interest, applications and candidate experience
Experience	Two people react differently to the same event	How do expectations, memories and context shape the experience?	Experience ratings, attention and later evaluation
Behaviour change	Someone endorses a goal but does not act	Which reactions, habits or settings support or block action?	Observable behaviour before and after an appropriate intervention

These are possible research situations, not reported Neurofactor treatment effects. They show what could be tested. Each of the five subseries examines its own questions, methods and limitations in depth.

Marketing: what a brand means

A campaign may build reach and familiarity without establishing associations that matter when consumers choose. Awareness may no longer be the problem; relevant differentiation may be. Research can explore which meanings a brand evokes, which are shared with competitors and which relate to preference. Then a change in messaging needs to be tested against preference or choice, not merely recall. Brand equity research has long separated familiarity from meaningful brand knowledge (Keller, 1993).

Sales: the meaning of purchase risk

The phrase 'too expensive' can mean exactly what it says: the buyer cannot afford the price. But it might also signal uncertain returns, expected internal resistance or concerns about implementation. It would be wrong to assume every price objection hides a psychological problem. Better research keeps these alternative explanations open and gathers evidence to distinguish them. Sometimes the solution is different communication; sometimes it is a trial, a product improvement or a simpler process.

Recruitment: the meaning of an employer promise

A job description may promise autonomy. One candidate hears 'trust and freedom'; another hears 'little support'. Rewriting a careers page does not demonstrate that the employer becomes more credible. The application process, managers' behaviour and existing employees' experiences matter too. A credible employer value proposition is grounded in what candidates can actually expect, not just attractive language.

Experience: the meaning of an encounter

A visitor who already knows a painting may arrive with familiarity and expectations. Another may be working hard to make sense of what they see. Research into aesthetic appreciation considers perception, knowledge, interpretation and emotion (Leder et al., 2004). That does not entitle us to infer an individual's exact thoughts from a single physiological signal. Where useful, combine observation and well-designed experience questions with appropriate measurements.

Behaviour change: the meaning of action

Someone may believe exercise is important but continue driving their usual route home. That does not necessarily indicate a negative attitude towards physical activity. Habit, opportunity, fatigue and the availability of alternatives may matter more at the moment of action. Behaviour change therefore begins with a diagnosis of the situation, not simply the message. Habit research explains how recurring contexts can trigger well-practised behaviour without each step being reconsidered (Wood & Rünger, 2016).

A frequent mistake is to treat a positive association as evidence that someone will buy or participate. That conclusion is premature. Price, availability, social expectations, skills, habits and competing goals all stand between evaluation and behaviour. Within the theory of planned behavior, Ajzen (1991) distinguished attitudes, subjective norms, perceived control, intentions and behaviour. Research on the intention-behaviour gap shows that plans often fail to translate into action (Sheeran & Webb, 2016).

That does not make association research irrelevant. It can illuminate an aspect of a decision that ordinary surveys or funnel reports do not explain. But that insight must be positioned correctly within a wider account of decision-making. A difference between what someone evaluates positively and what they actually choose is useful information; neither measurement is automatically the 'real truth'.

No single method captures every association. The research design must fit the question. Start by distinguishing content, evaluation, accessibility and behaviour.

Method	What it may help examine	What it cannot establish on its own
Free associations and interviews	Meanings people mention and their explanations	Every unspoken automatic reaction
Ratings and explicit judgments	Deliberate evaluations, expectations and stated reasons	That answers will map directly onto behaviour
Reaction-time tasks / IAT-type designs	Relative associative compatibility in a particular task	Secret thoughts, individual diagnoses or certain purchase probabilities
Eye tracking	Where and when visual attention is directed	Whether something is liked, trusted or believed without other evidence
EEG	Time-varying electrical brain activity and selected experimental responses	The exact words, personal memories or motives in someone's mind
Observation and behavioural data	What someone chooses or does under defined conditions	Which association caused the behaviour without further testing

The original Implicit Association Test (Greenwald, McGhee, & Schwartz, 1998) compares response times across category combinations. It may reveal information about relative associations under task conditions, not a hidden personal truth or a certain forecast for an individual. Reviews show that implicit and explicit measures may have different, limited and context-dependent relations with behavioural criteria (Gawronski & Bodenhausen, 2006; Kurdi et al., 2019).

A stronger study can combine sources when each answers a different question: associations and an observed choice, for example, or stated expectations and actual behaviour. Additional technology only improves a study if it contributes relevant, interpretable information.

An association map matters when it informs a decision. This process prevents the visualisation from becoming the end product.

1. Define the choice. What decision, experience or action must you understand? For whom, when and in what setting?
2. List competing explanations. Is attraction insufficient? Is there risk, weak differentiation, unconvincing evidence or a practical barrier? Several causes may coexist.
3. Choose appropriate measures. Measure associations, evaluations and behaviour separately where needed, and document what each measure can and cannot show.
4. Investigate meaningful differences. An overall average may conceal people with different motives or circumstances. Do not infer individual mental states from group-level data.
5. Formulate a specific hypothesis. For example: 'Showing how beginners are welcomed in the first lesson will increase trial bookings in this audience.' This is a prediction to test, not an established outcome.
6. Test the change. Where feasible, use an appropriate comparison, baseline and predefined success measure. Do not stop at message liking if the goal is enrolment.
7. Evaluate and refine. Did the association change? Did behaviour change? Could another mechanism explain the result? An inconclusive result is still informative.

The distinction matters: research can tell us what people associate, how they evaluate an offer and what they do. These observations may complement one another, but they are not interchangeable.

Return to the music school. Suppose interviews indicate that playing with other people is attractive, while some potential students fear falling behind. This is a fictional illustration, not a quantified research finding.

A superficial response is to run an even more enthusiastic campaign about making music together. An associative diagnosis suggests a more useful research question: can we keep that appealing future while making the uncertainty of the first lesson easier to resolve?

We might compare two versions of a trial-lesson offer. One highlights the joy of playing together. The other also shows how beginners are welcomed, what prior skills are not required and how participants meet the group. We would examine understanding, expectations and actual trial bookings. Even if one version performs better, further evidence may be needed to determine what caused the difference.

The intervention might also need to change the experience itself. If the lesson really is difficult for beginners, communication alone will not solve the underlying problem. The association can be a clue that something in the service needs to improve.

The central associations series introduces five more focused lines of inquiry. Marketing explores meaning, brand preference and differentiation. Sales examines purchase motives, resistance, evidence and delayed decisions. Recruitment focuses on employer perceptions, applications and credibility. Experience looks at the influence of expectations and context. Behaviour change links associations to habits, environments and practical opportunities for action.

Each subseries begins with its own independent introduction before moving into practical research questions. The central hub helps you choose a starting point. Individual articles remain readable on their own, including for visitors arriving from search engines.

For definitions and methodology, the existing knowledge base includes articles on associations and measuring associations. Lovable must verify those routes and publication states before displaying the links. Planned knowledge-base articles on association networks, congruence, implicit attitudes and associative choice models should not be linked until actually published.

Learning that an audience associates an offer with 'security' does not establish whether they will buy. The finding becomes useful when you identify what kind of security they mean, which alternatives they face, at what decision point it matters and how you can test your explanation.

The value of association research does not lie in a magical explanation of behaviour. It lies in better questions. Instead of assuming that the meaning an organisation tries to communicate is the meaning people receive, you examine what they actually associate. Then you test whether a change in communication, experience or environment changes a meaningful outcome.

Associations can illuminate attraction, resistance, differentiation, persuasion and behaviour change. They do not, by themselves, explain all behaviour. That combination of depth and restraint makes them useful for better decisions.