Source
- Ryan & Deci (2000), LIT-559, read in NOTE-442: the continuum (pp. 72–73) and the definitions of introjection, identification and integration.
- Deci & Ryan (2000), LIT-558, read in NOTE-439: the account of the self (p. 247), integration requiring autonomy support (p. 237), and Deci et al. (1994) as reported (pp. 237–238).
- Ryan & Deci (2006), LIT-560, read in NOTE-441: autonomy as “a matter of degree” (p. 1562) and SDT set beside the hierarchical accounts (pp. 1560–1561).
- Vansteenkiste, Ryan & Soenens (2020), LIT-561, read in NOTE-440: autonomy as “self-endorsed and authentic” volition, measured by self-report (p. 3).
What was actually shown
The continuum. Ryan and Deci (LIT-559) distinguish four kinds of extrinsic motivation by how far a regulation has been taken into the self. Introjection is “taking in a regulation but not fully accepting it as one’s own”, run by guilt and contingent self-esteem. Identification is “a conscious valuing” of the behaviour. Integration is identification “brought into congruence with one’s other values and needs” (pp. 72–73). Self-reported reasons for each correlate in a “quasi-simplex pattern” (Ryan & Connell 1989, p. 73): neighbouring styles correlate more than distant ones, which is the evidence offered for one underlying dimension of relative autonomy.
Integration differs from mere internalization. In Deci, Eghrari, Patrick and Leone (1994), as LIT-558 reports it (pp. 237–238), people in conditions with two or three supports (rationale, acknowledgement of feelings, choice) showed positive correlations between later free behaviour and reported valuing, enjoyment and freedom. People in conditions with one or none still internalized, but the correlations were negative: those who did more felt less free and enjoyed it less. The same amount of internalized behaviour came in two kinds. An identified value can also stay “compartmentalized (poorly integrated)” (p. 237).
Not order. SDT’s criterion is congruence with the whole of one’s values, “endorsed by the whole self” (LIT-560, p. 1560), not the level of the attitude that endorses. Heteronomy includes “inner impulses or demands” (p. 1561). Ryan and Deci cite the hierarchical accounts (Frankfurt 1971, Dworkin 1988) as reaching “conclusions similar to the phenomenologists” (pp. 1560–1561), but they make autonomy graded where those accounts make it a matter of whether an endorsing attitude exists.
The claim is the record’s synthesis of the four readings. None of them states it against hierarchical accounts in this form.
What this does not say
- It does not answer the regress. Ryan and Deci set it aside as practically finite (LIT-560, p. 1561), which is no answer to Velleman. “Integration with the self” either names a process specified without the agent’s participation, which could in principle be disowned, or presupposes the self it was meant to explain. Deci and Ryan’s “nascent self”, the organismic integrative tendency and intrinsic activity (LIT-558, p. 247), is their candidate for a stopping point; that it cannot be disowned is asserted, not shown.
- It does not say integration has been measured. “In some studies, identified, integrated, and intrinsic forms of regulation have been combined” (LIT-559, p. 73). The continuum rests on self-report.
- It does not say manipulated integration is autonomous or not. That question is THEORY-040’s, and THEORY-047 takes up the historical half.
Connections
- THEORY-029 (no higher-order attitude makes a motive one’s own by its order alone). This account agrees and supplies the empirical alternative: ownership by graded integration. It is not declared as extending THEORY-029, because that account is a negative claim about hierarchical theories and this one is a positive claim with a different criterion.
- Frankfurt (LIT-293, NOTE-264). His externalized desires, ours “in a gross literal sense” (p. 174), are SDT’s introjects, “within the person, but still relatively external to the self” (LIT-558, p. 235). His identification can be “weary resignation” (p. 173), which SDT would score as controlled. His free will is “volitional unanimity” and his love is stipulated unambivalent, where SDT grades.
- The autonomy surveys (LIT-296, NOTE-267; LIT-297, NOTE-266). In their classes, this criterion is a coherentist one close to the “integration” account LIT-296 attributes to Arpaly and Schroeder, with a historical element (THEORY-047) and a substantive one, since some values are said to be “not integrateable” (LIT-558, p. 246).
- THEORY-047 extends this account to the conditions that produce integration.
THEORY-120, Frankfurt’s hierarchical account, read first-hand from 1971 (LIT-692), is this account’s rival. On Frankfurt’s account, a desire is the agent’s own once a second-order volition endorses it. On this one, autonomy is graded integration, and an endorsement that is not integrated is introjection. The two give opposite verdicts on resigned identification.
Note of 2026-10-29: what holds it back
Kept Proposed by the eighth status review. No reading separates integrated from identified regulation: the SDT sources say the forms are usually combined, and the self-concordance study (NOTE-448) pools identified and intrinsic reasons. Nor does the record hold a case in which a motive endorsed at a higher order behaves like a controlled one. THEORY-074’s overlap index (NOTE-469) is a candidate measure of integration apart from endorsement, untested as one.
Note of 2026-11-02: the evidence its promote_when asks for
In the exchange “Assess SDT Coverage” (2026-11-02), A2 named Howard, Gagné and Bureau 2017 (LIT-1996), a meta-analysis of the continuum of self-determined motivation, as bearing on whether integrated regulation is separable from identified regulation. A2 says it does not find integrated regulation as a distinct factor. That is unverified until the paper is read. Deci et al. 1994 (LIT-1973), which this theory and THEORY-047 hold second-hand, is registered too.
The status is unchanged.
Note of 2026-11-03: Howard et al., second-hand
Howard et al. 2017 is Unfetched. Ryan et al. 2022 (NOTE-1500) report it: integrated regulation was not separated from identified, partly because few studies measured integration, and Van den Broeck et al. 2021 found the same at work. So the promote_when’s first clause stays unmet, now on two reports. The causality orientations paper (NOTE-1531, p. 111), where inner “I have to” events count as controls, is an early SDT statement of the claim’s “not order” half.
The status is unchanged.