Yoked Control Design¶

An experimental pairing in which one participant's response determines an event sequence that a partner receives without controlling it, matching exposure while comparing contingency.

Version: v1 · 2026-10-03 · History
Domain-specific #: 13700
Domain group: Social Sciences
Origin domain: Psychology & Behavioral Sciences
Subdomains: Experimental Psychology, Behavioral Research → Psychology & Behavioral Sciences
Aliases: Yoked comparison design

Core Idea¶

A yoked control design links two experimental units through a designated event sequence. One unit's response or choice determines that sequence; a paired unit receives a matching sequence, but its own corresponding response does not determine it. The construction tries to distinguish effects of having control over the event from effects of being exposed to the event pattern. In the historical contingency form, the yoked participant is not behaviorless: the point is that its behavior does not alter the matched outcome stream. Maier and Seligman's first-person historical review describes exactly this contrast, and Murayama and colleagues use the same source-choice/yoked-exposure structure in a human word-list study.^[1]^[2]

Yoking matches a specified exposure; it does not make two participants interchangeable or guarantee that literally one factor differs. Assignment, baseline differences, the experience of deciding, attention and context can remain unequal. The inference depends on which exposure dimensions were matched, which outcome was measured and which competing explanations were checked. The name “control” is also role-relative: in the learned-helplessness comparison, a yoked uncontrollable condition was often the focal experimental condition, not simply a neutral baseline.^[1]^[2]

Structural Signature¶

Sig role-phrases: linked pair and assignment → response-contingent source sequence → yoked matched exposure without own contingency → outcome contrast with inferential boundary.

Linked pair and assignment. Two units are paired so that one unit's event sequence supplies the other's designated exposure. How the units are assigned matters for what causal inference can follow. A superficially matched pair is not automatically balanced on prior traits.^[1]^[2]
Response-contingent source sequence. The source unit can influence an event pattern or its endpoint by a relevant action. In the human list example, one participant chooses when the stream of words stops. The exact kind of action is an accent; its influence on the designated sequence is core.^[1]^[2]
Yoked matched exposure without own contingency. The partner receives the specified event sequence produced by the source unit, yet its corresponding behavior does not change that sequence. Which dimensions are identical—items, order, amount or timing—must be stated rather than presumed.^[1]^[2]
Outcome contrast with inferential boundary. The study measures an outcome across conditions to ask whether response contingency matters when designated exposure has been aligned. The contrast is informative, not self-authenticating: selection effects and differences tied to decision-making can survive the match.^[1]^[2]

What It Is Not¶

It is not proof that only one factor differs. The matched stream removes one obvious explanation based on the designated exposure amount or pattern. It does not equalize every participant characteristic, expectation, or cognitive process. Maier and Seligman themselves warn that an earlier yoked comparison could be confounded by how units entered the active versus yoked role. Murayama and colleagues discuss whether making a stop decision uses attention in a way passive exposure does not.^[1]^[2]

It is not generic matched-stimulus pairing. If two people merely judge the same fixed letters but neither person's response determines what the other receives, the classical response-contingency structure is absent. The frozen seed's name-letter extension is held as a possibly related broader pairing; Nuttin's inspected publisher abstract supports the name-letter effect, not the exact response-determined yoking mechanics required here. This draft does not silently accept that variant as the same identity.^[3]

It is not a requirement that the yoked partner make no response. Their behavior can be observed or measured; it is noncontingent for the specific paired event stream. Nor must every study add a third no-exposure condition. Such a baseline can help interpret outcomes, as in the historical triadic comparison, but the dyadic yoking relation is the defining match.^[1]

It is not interchangeable with blocking. Blocking groups units by prior characteristics, while yoking links an experienced sequence from one unit to another. Both may occur in one experiment, but matching an event stream is a different operation from partitioning units by a covariate. Likewise, a “control sample” can be part of the comparison without defining which arm is the scientific focus.

Scope of Application¶

In the historical behavioral contingency comparison, the original investigators retrospectively describe a condition whose participants could affect their outcomes and a yoked condition receiving the same designated outcome pattern without that response contingency. A separate no-exposure group supplied another comparison. The scientific question concerned later behavior under a new task. Here the historical case is used solely to analyze the matching logic and its inferential limits; it is not an experimental procedure to reproduce.^[1]

In a distinct human learning study, Murayama, Blake, Kerr and Castel compared participants who could choose when to stop seeing a sequence of words with yoked partners exposed to the same word order and length without making that stopping choice. Their Experiment 3 measured subsequent recall. The participant who stopped the list determined the exposure endpoint; the paired participant's own choice did not. The match helps distinguish word-count exposure from agency/decision effects, although it cannot by itself separate every cognitive difference associated with deciding.^[2]^[4]

Other experiments may use yoked event streams, but membership requires more than the word “yoked.” The source unit's behavior must determine a specified exposure that is matched to a partner whose behavior is noncontingent for that stream, and a measured contrast must answer a stated question. The reference source's access limits and study-specific assignment details remain part of later promotion review.^[1]^[2]

Clarity¶

The design clarifies a causal question that an ordinary exposed-versus-unexposed comparison cannot answer cleanly: did the event itself matter, or did the ability to influence it matter? Pairing the experienced stream narrows that ambiguity. It does not eliminate all confounding, so the right statement is “the specified event sequence was aligned,” not “every other variable was controlled.”^[1]

It also clarifies why the yoked participant may be the experimental subject. In a study of uncontrollability, the unusual condition is the lack of contingency despite matched exposure; the active source group may serve as one control for exposure. The term “yoked control” names a comparison device, not an immutable label for which group is experimental. The original review explicitly discusses this role reversal.^[1]

Manages Complexity¶

A response-contingency experiment can confound event amount, event sequence and control. Yoking turns one variable—the specified exposure stream—into a shared record for a pair. Instead of reasoning over unrelated histories, the investigator can compare matched histories that differ in whether the participant's own action influenced them. This compresses one major source of variation while preserving a contrast of agency.^[1]^[2]

That compression has a cost: the event-producing participant and the yoked partner may differ in attention, effort, expectation or preexisting traits, and pair-specific histories may not be representative. Matching outcomes does not match all internal processes. Analysis therefore still needs assignment checks and alternative explanations; in the human word-list example, the act of deciding when to stop can itself consume attention, a possible mechanism rather than noise to be wished away.^[1]^[2]

Abstract Reasoning¶

First identify the source-controlled event stream and specify exactly which parts of it the partner receives—amount, order, timing or stopping point. Then identify the response whose contingency is present for one unit and absent for the other. State which outcome is compared and what exposure-based alternative the match is intended to weaken. If the partner can alter the matched stream in the same way, or if the two streams are not in fact matched on the claimed dimension, the yoked-control inference fails.^[1]^[2]

Next test the residual: Were units assigned comparably? Does choosing the stream also change attention or strategy? Could a measured outcome depend on such differences rather than the hypothesized contingency alone? The design makes these questions sharper but does not answer them automatically. A third group, additional measures or another design component can address a remaining alternative, but none is a constitutive part of the two-unit yoking relation.^[1]^[2]

Knowledge Transfer¶

The pairing logic transfers literally between the historical contingency study and the human list-stop study: a participant's action determines the event history, a partner receives the matched history without controlling it, and a later outcome is contrasted. The event substrate changes profoundly, but the causal-design roles remain recognizable. The lesson is not that psychological mechanisms transfer from one study to the other; only the experimental architecture does.^[1]^[2]

A general idea of “matching inputs while varying agency” might travel further, but this named design remains tied to intervention assignment, event-stream matching and outcome measurement. That broader matching/agency skeleton is a future-prime question. Live prime Experimental Design already supplies the tested strict genus; no additional speculative parent is needed.

Examples¶

Historical response-contingency pairing¶

Maier and Seligman's first-person review describes a controllable-event condition paired with a yoked condition receiving the same designated event pattern while lacking the corresponding response-outcome contingency. A separate no-exposure group helped contextualize a later behavior measure. The example establishes the comparison architecture only; this entry intentionally omits stimulus apparatus, settings, durations and operational procedure. The review also documents why equal designated exposure alone cannot excuse weak assignment or eliminate every rival explanation.^[1]

Mapped back: Linked pair and assignment → a source participant paired with a yoked participant; response-contingent source sequence → the source participant's action determined the relevant event pattern; yoked matched exposure without own contingency → partner received that pattern but its corresponding action did not change it; outcome contrast with inferential boundary → later behavior compared with attention to assignment and alternative causes.

Human word-list stopping and recall¶

In Murayama and colleagues' Experiment 3, a participant in the stop condition decided when a word sequence ended. A yoked participant received the same sequence and number of words without making that stopping decision; the study then compared recall. The original report describes pair-level matching and pseudorandom assignment. This human case is not an aversive-event analogue in psychology; it instantiates the same experimental pairing logic in a different task. The publisher PDF's detailed method was available through indexed text, while direct full-PDF rendering was unavailable to this author and remains a pre-promotion source check.^[2]^[4]

Mapped back: Linked pair and assignment → stop participant linked to a pseudorandomly assigned yoked partner; response-contingent source sequence → stop choice determines the word-list endpoint; yoked matched exposure without own contingency → partner sees the matched word order and count without determining the endpoint; outcome contrast with inferential boundary → recall comparison, with decision-related attention remaining a possible process difference.

Structural Tensions¶

T1: Match exposure versus isolate agency. Matching the sequence removes a direct difference in designated exposure, strengthening a contrast about response contingency. Yet the source participant must make the choice that defines the sequence, while the yoked partner does not; engagement, attention and expectation may differ. Trying to eliminate every such difference can erase the agency contrast itself, while ignoring them makes the causal conclusion too strong. Diagnostic: Which dimensions of exposure were actually matched, and which cognitive or subject-selection differences remain relevant to the measured outcome?^[1]^[2]

Structural–Framed Character¶

Yoked Control Design lies in the mixed structural–framed region, leaning toward the research-practice frame. Evaluative weight: matching exposure is a tool for stronger inference, but design membership does not certify unbiased causality. Human-practice dependence: an investigator deliberately links units, assigns conditions and decides what to measure; absent those choices no yoked experiment exists. Institutional origin: the historically named studies are evidence of use, not authorities that define every future pairing. Vocabulary travel: “yoked” can mean merely paired or mechanically linked; only the response-determined event match and noncontingent partner meet this identity. Import versus recognition: importing the design to a new question requires finding a genuine source-controlled sequence and a partner who receives it without controlling it, plus an outcome contrast—not just labeling two participants a pair.^[1]^[2]

Its character: a reusable experimental structure across unlike research materials, but domain-specific because deliberate assignment, exposure matching and causal-outcome interpretation are constitutive. The general idea of matching experienced input while varying agency would require separate prime-level adjudication.

Structural Core vs. Domain Accent¶

The core is the pair link, source response-contingent event stream, partner's matched noncontingent exposure and measured contrast with an explicit inferential limit. Historical aversive events and human word lists are accents; neither material is a required part of the design. A no-exposure third group is useful in one research question but not the minimal dyadic structure. A merely shared static stimulus set without a source participant's controlling response would be a broader matched-stimulus design, not the same admitted identity.^[1]^[2]

The portable skeleton “same exposure, different agency” is a future-prime question, not a license to merge this method with every paired comparison. The actual proposed parent is live prime Experimental Design, because the node is a deliberately structured comparison of assigned conditions and measured outcomes. Live Blocking in Experimental Design can also match units but does not entail stream copying from one participant's action; no strict edge to Blocking is asserted.

This entry is a kind of Experimental Design.

One strict upward edge to Experimental Design is proposed. Control Sample may be involved, but the yoked arm's scientific role is question-dependent and cannot be equated with a generic control sample. Comparison describes the broad relation between conditions without the intervention architecture. Blocking in Experimental Design addresses covariate grouping rather than a response-generated shared event stream. The Nuttin name-letter judgment example remains a separate broader-pairing identity hold until its yoking mechanics and scope are independently established.

Relationships to Other Abstractions¶

Current abstraction Yoked Control Design Domain-specific

Parents (1) — more general patterns this builds on

Yoked Control Design is a kind of Experimental Design Prime

Yoked pairing is a specific planned intervention-and-outcome experimental design.

Hierarchy paths (2) — routes to 1 parentless root

Yoked Control Design → Experimental Design → Control Sample → Comparison → Self Checking

Show alternative path (1)

Neighborhood in Abstraction Space¶

Yoked Control Design sits in a moderately populated region (56^th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Language, Mind & Meaning-Making (57 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With¶

Only-one-factor certainty: designated exposure matching does not equalize every participant or process variable.^[1]^[2]
No behavior by the yoked unit: only the matched event stream is noncontingent on its relevant behavior.^[1]
Mandatory third arm: a no-exposure comparison can improve interpretation but is not part of the dyadic yoking rule.^[1]
Generic shared-stimulus judgment: seeing the same fixed items is insufficient without a source participant's response determining the paired stream.^[3]
The particular learned-helplessness claim: the yoked pairing is an experimental architecture; it does not itself prove any one psychological theory.^[1]

References¶

[1] Steven F. Maier and Martin E. P. Seligman, “Learned Helplessness: Theory and Evidence”, Journal of Experimental Psychology: General 105(1), 1976, especially first-person triadic/yoked design discussion, PDF pp. 3–4, and assignment-confound caution, PDF p. 12. Cited solely for historical conceptual design and inferential limits, not procedural reproduction. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y

[2] Kou Murayama, Adam B. Blake, Tyson Kerr and Alan D. Castel, “When Enough Is Not Enough: Information Overload and Metacognitive Decisions to Stop Studying Information”, Journal of Experimental Psychology: Learning, Memory, and Cognition 42(6), 2016, Experiment 3. Publisher-original PDF method was checked via indexed original text; direct PDF reader returned a wrapper, and the author manuscript host denied direct access. Detailed method needs direct full-text recheck before promotion. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s

[3] Joseph M. Nuttin, “Narcissism Beyond Gestalt and Awareness: The Name Letter Effect”, European Journal of Social Psychology 15(3), 1985, publisher abstract only; does not establish the response-contingent yoking mechanism for this entry. registry ↩a ↩b

[4] PubMed bibliographic record for Murayama et al., PMID 26595067, DOI 10.1037/xlm0000213, original-article metadata and abstract. registry ↩a ↩b