Contribution
It updates Aron et al.’s 2013 handbook survey of research on inclusion of other in the self, for long-term romantic pairs. It sorts studies into measurement, predictors and outcomes, and adds two newer theoretical strands. Its thesis is that pair-bonding and IOS are two faces of one thing: the desire to include the other drives the bond, and “the pair bond itself is an enduring feature of having successfully included the other in the self” (p. 1).
Key insight
On this model, including a partner in the self is a cognitive fact before it is a feeling. The partner’s “resources, perspectives, and identities” are experienced “as to some extent one’s own”, so that “the cognitive construction of the other merges with the cognitive construction of the self, and that person’s outcomes are shared” (p. 1, citing Aron et al. 1991 and Mashek and Aron 2004). The evidence the review leans on is of that kind: people process a partner’s traits, pain and comparisons as they process their own.
Assumptions
- Self-expansion motivation. People are motivated to expand their efficacy, and including a partner’s resources does it (p. 1). This is assumed, not tested here.
- The IOS scale measures inclusion. It is called the most common and “arguably, the most face-valid” measure (p. 2), but its validation is as a measure of closeness.
- Pair-bonding and IOS are near-identical constructs (p. 1). This makes “IOS predicts pair-bonding” close to definitional.
Key results
- Measurement (p. 2).
- The IOS scale (Aron et al. 1992): seven pairs of circles, each overlapping more than the last; respondents pick the pair that best portrays the relationship. The original validation “captured aspects of both feeling close and behaving close, and correlated strongly with more complex, multi-item measures of closeness and intimacy”.
- It works across cultures and has a continuous web version (Le et al. 2007). The originating paper has “over 3,800 citations”.
- Gächter et al. (2015) found it “a psychologically meaningful and highly reliable measure of the subjective closeness of relationships”.
- Other measures used to validate it include “the implicit me-not-me task (Aron et al., 1991, Study 3)”. Facebook-profile overlap correlates with IOS and with commitment (Castañeda et al. 2015).
- Predictors (p. 2).
- Escalating reciprocal self-disclosure with a stranger (Aron et al. 1997).
- Shared exciting activities in marriage at year 7 predicted satisfaction at year 16, mediated by IOS change (Tsapelas et al. 2009).
- Induced positive affect (Waugh and Fredrickson 2006); humour (Treger et al. 2013).
- A perspective-taking induction before conflict buffered the effect of the partner’s attachment avoidance on post-conflict overlap (Bernstein et al. 2015).
- Outcomes, from the 2013 review (pp. 2–3).
- Me/not-me: “one’s ability to process information about the self on a particular trait is slowed to the extent that the other is dissimilar on that trait (Aron et al., 1991)”.
- Self–close-other confusions in recalling trait adjectives exceed self–non-close-other confusions (Mashek et al. 2003).
- More plural pronouns go with more inclusion (Agnew et al. 1998).
- Pain of self and close other processed alike, not a stranger’s (Cheng et al. 2010). A close other’s success celebrated, not felt as threat (Gardner et al. 2002).
- Outcomes, newer (p. 3).
- Partners of high-overlap individuals report more support and commitment, but only when those individuals are high in self-determination (Weinstein et al. 2016).
- Couple identity fusion: a balanced blend of both selves goes with less vigilance for threat and more constructive coping; an imbalanced one, with less (Walsh and Neff 2018).
- Well-being is best when actual and ideal IOS match, whatever the level (Frost and Forrester 2013).
- Reminded of a partner’s inferiority, high-overlap people keep a positive global view of the partner; low-overlap people do not (Thai and Lockwood 2015).
- A spouse’s sleep is worse after the patient’s daily knee pain, more so in high-IOS couples (Martire et al. 2013).
- Honour-endorsing men high in IOS with their wife or partner were the most aggressive toward a hypothetical moral violation (Benavidez et al. 2016).
- After a break-up, people keep attributes gained from the partner that they worked to maintain. Attributes that conflict with prior beliefs lower self-concept clarity (Slotter et al. 2014).
- Theory (p. 4).
- The two-dimensional model of relational self-change (Mattingly et al. 2014; McIntyre et al. 2015). Gaining or losing content, positive or negative, gives four processes: expansion, pruning, contraction and adulteration. The improving ones go with love, satisfaction, commitment and willingness to sacrifice. The degrading ones go with infidelity, revenge and attention to alternatives.
- IOS-perceived (Tomlinson and Aron 2013): the IOS scale answered as if one were the partner. Perceived partner satisfaction leads to IOS-perceived, which leads to one’s own IOS.
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | In close relationships, partners treat each other’s resources, perspectives and identities as partly their own | moderate as reported: rests on Aron et al. 1991 and the 2013 review, summarized in two sentences | p. 1 |
| C2 | The IOS scale is a reliable, face-valid measure of subjective closeness | moderate to strong: validation studies cited, including an independent one (Gächter et al. 2015) | p. 2 |
| C3 | Inclusion shows up in cognition: slowed self-judgements on traits where the partner differs, self–partner memory confusions | moderate as reported: two studies cited without numbers | pp. 2–3 |
| C4 | Greater overlap goes with treating the partner’s outcomes and standing as one’s own, for good (support, positive regard) and ill (sleep loss, honour aggression) | weak to moderate: single correlational studies, reported without numbers | p. 3 |
| C5 | Pair-bonding and IOS are complementary, nearly the same construct | weak: stated, and close to definitional | pp. 1, 4 |
Concepts
- inclusion of other in the self (IOS): treating a close other’s resources, perspectives and identities as partly one’s own.
- IOS scale: a single item of seven increasingly overlapping circle pairs.
- me/not-me task: reaction times for judging whether traits describe oneself, slowed where self and partner differ.
- two-dimensional model of relational self-change: self-expansion, self-pruning, self-contraction and self-adulteration.
- IOS-perceived: one’s estimate of how far the partner includes oneself.
- couple identity fusion: a balanced or imbalanced blend of both partners’ selves in the couple identity.
Connections
- Aron et al. 1991 (LIT-1706). The review gives the paper’s conclusion and its me/not-me result in one sentence each. The SEP entry, read the same day, quotes a line attributed to “Aron et al. 1980: 242” that is probably from it.
- Cialdini et al. and Batson et al. 1997 (LIT-1728, LIT-1700). Both used the IOS scale as their overlap measure. The review does not mention the altruism dispute.
- Identity fusion (NOTE-1256). Swann’s fusion item was adapted from this scale. Walsh and Neff carry fusion into couples and find balance matters, not only the amount of overlap.
- León’s personalized “we” (THEORY-1115). Walsh and Neff’s result, that a balanced couple identity beats one where either self dominates, fits León’s claim that the “we” is healthier the more differentiated the “I”. Weinstein et al.’s self-determination moderation points the same way. An inference.
Bearing on the record
- The triviality objection (CLAIM-492). This is the record’s best
material on the objection’s demand, though it does not meet it. The
following are inferences; the review does not discuss motivation for
others’ sake.
- The IOS scale itself is validated as closeness, “feeling close and behaving close”. As a criterion of the self’s extent it is not clearly apart from caring.
- The cognitive signs are another matter. Slower self-judgements on traits where the partner differs, and self–partner memory confusions, are effects on the processing of one’s own self-concept. Caring about someone does not obviously predict them. They are candidates for a criterion of extension that can come apart from caring.
- Some outcomes separate inclusion from benevolence. Honour-endorsing men high in inclusion are more aggressive toward a partner’s moral violation. Treating a partner’s conduct as touching one’s own honour is what self-extension predicts, and what concern for her welfare does not. The review reports this as one correlational study.
- The owner’s thesis (CLAIM-491). Self-expansion is cited there as an “inclusive self”. The review confirms that the model claims this: partners’ “outcomes are shared” (p. 1). But the model’s motive is expanding one’s own efficacy by acquiring the partner’s resources. It fits the thesis’s “larger self”, but as acquisition, not as concern for the partner. An inference.
- Batson’s objection (CLAIM-494). Not addressed.
- A dated note to CLAIM-492 is proposed:
Note of 2026-10-29: what IOS measures. Branand, Mashek and Aron (NOTE-1276) report that the IOS scale is validated as a measure of subjective closeness, so as a criterion of the self’s extent it is not clearly apart from caring. The self-expansion literature also has indirect signs of inclusion: slowed self-judgements on traits where a partner differs (Aron et al. 1991), self–partner memory confusions (Mashek et al. 2003). Those concern the processing of one’s own self-concept, and caring alone does not predict them. A test of the thesis could use them in place of the circles.
- No THEORY filed. No ML instruction.
Limitations
- A mini review by the model’s originator’s group. It is selective, gives no effect sizes, and does not discuss criticism of the model.
- Romantic pairs only. Nothing on groups, strangers or helping.
- Most outcome studies are correlational and single.
- Aron et al. 1991 is reported in two sentences. Its resource allocation, perspective and memory studies are not described.
- Typos and slips.
- “valance” for “valence” (p. 4).
- The abstract credits the two-dimensional model to McIntyre et al. 2015; the text credits it to Mattingly et al. 2014.
- “self-degradation” (p. 4) for “self-adulteration”, the term used elsewhere.
- The model is called “relational self-change” in the text and “relationship self-change” in the Mattingly et al. title.
- A missing space in “(Gächter et al., 2015).Offering” (p. 2).
Open questions
- Do the cognitive signs of inclusion (me/not-me latency, self–other confusions) predict helping a partner when closeness or empathic concern is held fixed? That would test whether inclusion, not caring, carries the motive.
- Is inclusion with a group measured the same way and with the same cognitive signs? The review is silent; the fusion literature claims so.