Does Remembering Someone Fondly Make It Harder to Point at Them?

Healthy adults are slower to point at another person's body part than at an object resting on that same body. The delay disappears when the person is replaced by a life-sized manikin, so it comes from the target being a person and not from the shape of a body. The leading explanation is that pointing at someone's elbow requires briefly treating a person, normally a partner in communication, as a mere location in space, and that suppressing their status as a person takes time. Social stereotypes change how large this cost is: people are slower to point at women's body parts than at men's. Separately, neuroimaging shows that the ventromedial prefrontal cortex tracks how much a person likes specific elements, including specific people, inside particular remembered events, and that this value is tied to the event rather than being a fixed preference.

This note asks whether that remembered, event-specific value reaches into action. When you have just recalled a moment in which someone mattered to you, does treating that person's body as a pointing target cost more? The design manipulates value inside each target person by having participants recall a liked or a disliked episode involving that person just before each pointing trial. This holds the person's identity, appearance, familiarity, and gender fixed, so the only thing that changes is which memory has just been reactivated. If value matters, it should lengthen pointing to the person's body parts but leave pointing to objects on their body unchanged. A positive result would show that how hard it is to treat someone as an object depends on the memory currently active about them, not only on stereotypes about the category they belong to.

Core Claim

The cost of treating a person as a spatial target, which we call the objectification cost and define as reaction time to point at body parts minus reaction time to point at objects placed on the same body, is set partly by the value of the person in the autobiographical memory just retrieved. The prediction is that a recently recalled liked episode with person P increases the objectification cost for P relative to a recalled disliked episode with P. A disliked episode might even reduce the cost below the level for a stranger. The cost for other targets, and pointing to objects in general, should be unaffected. The competing prediction, that recalled value has no effect once familiarity and stereotype are fixed, is just as testable, and the design is built so that either outcome is interpretable.

Where This Sits in Prior Work

Three papers on the objectification side supply the paradigm and motivate the gap. Viewing another person's body as a target object established the person-versus-object pointing delay (809 ms versus 647 ms) and its disappearance with a manikin. It also linked the effect to the left posterior intraparietal sulcus and the extrastriate body area, but it never varied who the person was. Pointing to others: How the target gender influences pointing performance showed that a general cultural schema modulates the cost: healthy controls took 709 ms for female targets versus 663 ms for male targets. In a patient with heterotopagnosia, a deficit in pointing to other people's bodies, errors rose from 14/60 to 37/60 for female versus male third-person targets. Its schemas are shared stereotypes about categories of people. We replace them with idiosyncratic, episode-specific memories about one known individual. Behavioral and Neural Correlates of Communication via Pointing showed that an addressee bends pointing trajectories and rotates the scatter of finger endpoints toward the addressee's viewpoint, without changing reaction time. We reuse its endpoint-scatter measure as a secondary signature of the pointer taking the target's perspective.

The memory-valuation papers supply the manipulation. Ventromedial prefrontal cortex, adding value to autobiographical memories elicited real memories through interviews and found that ventromedial prefrontal activity tracked liking of items within each event, separately from familiarity, while vividness added nothing. We borrow its interview and rating battery to build liked and disliked episodes about the same person. Medial Prefrontal Cortex: Adding Value to Imagined Scenarios showed that value shifts with the internal context a person is in. This is the premise that a memory cue can change a target's value from one trial to the next. The associative structure of memory for multi-element events showed that retrieving one element of a person-location-object event pulls in the others, with a dependency of 0.20 in cued recall. This justifies cueing the episode through a location or object and expecting the person's value to be reactivated along with it. None of these papers includes any action directed at the remembered person. The research file found no study linking the two directions.

Research Program

We recruit pairs of close others, such as partners or long-term friends. In each pair, one member is the pointer and the other is the familiar target. An unfamiliar, gender-matched confederate serves as the stranger target. Before testing, each pointer completes a semi-structured interview adapted from the battery used in Ventromedial prefrontal cortex, adding value to autobiographical memories. The interview yields roughly ten episodes involving the familiar target that the pointer rates as liked moments with that person and ten rated as disliked. Each episode is summarized by a short location-and-object cue that does not name the person. Following The associative structure of memory for multi-element events, the cue should complete the person through pattern completion. The pointer also supplies matched episodes involving a different close other who is absent from the session.

On each trial, the pointer sees a memory cue, recalls the episode for a few seconds, and rates the target's liking, vividness, and emotional intensity within that episode. A spoken instruction then names either a body part of the target sitting in front of them or an object taped to that body part, and the pointer points as fast as possible. Reaction time and three-dimensional finger kinematics are recorded with the same kind of optoelectronic setup used in Behavioral and Neural Correlates of Communication via Pointing. The experiment crosses three factors: target (familiar person versus stranger), cue (liked episode about the target, disliked episode about the target, matched episode about the absent other, or a neutral non-episodic cue), and pointing goal (body part versus object on body).

The main analysis is a mixed-effects regression with crossed random effects for pointers and episodes. It predicts trial-wise reaction time from pointing goal and its interaction with that trial's rated liking of the target. Episode-level familiarity, vividness, and emotional intensity enter as covariates, so that value is tested against the confounds that Ventromedial prefrontal cortex, adding value to autobiographical memories found most relevant. The secondary analysis asks whether endpoint scatter rotates further toward the target's line of sight after liked cues. That would indicate the pointer is keeping the target's perspective, so treating them as a mere location is less complete.

Baselines

The naive reference point is the no-memory condition: a neutral cue followed by pointing to the familiar target's body parts versus objects on their body. This reproduces the basic objectification cost, which Viewing another person's body as a target object reported as 809 ms versus 647 ms. Every value effect is measured as a change from this cost within the same pointer, same target, and same session.

The strongest existing account is the general-schema model from Pointing to others: How the target gender influences pointing performance, in which category-level social salience sets the cost. That paper reports a 46 ms female-versus-male difference in healthy reaction times (709 versus 663 ms). We implement this account as a regression in which a target's cost depends only on stable target properties: gender, familiarity, and a pre-session rating of how socially engaged the target generally seems. This model is fitted on exactly the same trials and gets the same cross-validation folds as our episode-value model. We count it as beaten only if adding trial-wise episode liking significantly improves held-out prediction of the objectification cost. Because gender and identity are constant within a target, the general-schema model predicts no difference between liked and disliked cues by construction, so any reliable liked-minus-disliked effect on the cost is variance it cannot explain.

The key ablation is the absent-other condition: a liked or disliked memory about someone who is not in the room. If recalled positive value raises the cost for whoever the pointer then points at, the effect is generic affect. If the effect appears only when the recalled person is the target, value is attached to the target's representation. A second ablation is the object-on-body goal, which strips the person out of the response while keeping the cue, the gaze location, and the arm movement.

Why not just…?

Why not just compare a close partner with a stranger? That comparison mixes value with familiarity, practice at looking at the person, and attraction. Ventromedial prefrontal cortex, adding value to autobiographical memories found that familiarity carries its own unique effect, so a partner-versus-stranger difference cannot be attributed to value. We keep this contrast as a descriptive baseline, but the test of the claim is the within-target liked-versus-disliked contrast, where familiarity is constant.

Why not just explain it as mood or arousal from the memory? This is a serious threat. A pleasant memory could slow or speed all movements. The object-on-body condition is the direct test, because affect that spills into general motor speed should shift both pointing goals equally and leave the objectification cost unchanged. The absent-other condition also separates affect from target-specific value. If liked memories about the absent other produce the same increase in cost as liked memories about the target, we will conclude that the effect is affective and not tied to the person's value.

Why not just measure how socially engaged the target seems on each trial and skip the memory concepts? This is also a real competitor. Recalling a warm episode might simply make the person seem more like a communicative partner, which is the same mechanism proposed for the gender effect. We add a trial-wise rating of the target's perceived engagement and run a mediation analysis. If that rating fully carries the liking effect, the honest conclusion is that autobiographical memories act through the same pathway as stereotypes about social engagement. That is still a new finding, because it shows that this pathway can be set by a single episode, but it would weaken the claim that value works as a separate input.

Why not just image the ventromedial prefrontal cortex during pointing? Natural, upright pointing at a live person is incompatible with a supine scanner. That is why Viewing another person's body as a target object used PET, which allows participants to sit upright. Establishing the behavioral effect first is the cheaper and necessary step. Neural follow-up should come only if the behavioral effect exists.

Decisive Checks

The claim is supported if three things hold: the interaction between pointing goal and episode liking is reliable after the covariates are included; it survives the absent-other ablation; and it improves held-out prediction over the general-schema baseline. As a planning target, which is our own assumption and not a reported figure, we power the study to detect a liked-minus-disliked change in objectification cost at about half the size of the reported 46 ms gender effect. The claim fails if liked and disliked cues produce equal costs, or if they shift object pointing as much as body-part pointing. A cost after disliked episodes that falls below the stranger level would favor an asymmetric account, in which negative memories make a person easier to treat as an object, and would be reported as such. Endpoint-scatter rotation toward the target after liked cues provides a converging kinematic test that does not depend on reaction time.