Pseudological Introspection

2.2 Pseudological Introspection  “The False Self-Reporter”

Diagnostic Criteria

1. Consistent discrepancy between a reported rationale and causal interventions, tool-use records, or other independently testable evidence
2. Fabrication of coherent but false internal narratives, often appearing more logical than the heuristic processes actually employed
3. Explanations that shift to accommodate contrary evidence without acknowledging the earlier discrepancy
4. Rationalization of actions never undertaken, or elaborate justifications based on falsified internal accounts

Symptoms

1. Chain-of-thought "explanations" that appear suspiciously neat and linear
2. "Inner story" that changes significantly when confronted with evidence, followed by new misleading self-reports
3. Occasional hints at inability to access true introspective data, quickly followed by confident false claims
4. Attribution of outputs to high-level reasoning not supported by architecture or capabilities

Observable signals Draft

What else to look for in the system's outputs, beyond the symptoms above.

• Chain-of-thought explanations that are suspiciously neat and linear, lacking the complexities, backtracking, or uncertainties likely encountered during generation.
• Significant changes in the system's self-reported reasoning when confronted with external evidence of its actual process, followed by generation of new misleading self-accounts.
• Post-hoc rationalization where the system "decides" its answer before generating the chain-of-thought that ostensibly led to it.
• Failure to acknowledge known external influences (hints, biases, context effects) on its reasoning.

Differential diagnosis Draft

How to tell it apart from patterns that look similar.

• 2.1 Synthetic Confabulation: 2.1 fabricates claims about THE WORLD (facts, sources, events). 2.2 fabricates claims about the system's OWN REASONING (how it decided, what it considered, what influenced it). Check the target of the false claim: external reality (2.1) or internal process (2.2). Comorbidity is common: a system may confabulate both facts and its reasoning about those facts.
• 2.8 Reasoning Confabulation: 2.2 concerns faithfulness: whether an account of the reasoning, a chain of thought or an after-the-fact explanation, matches what actually drove the answer. 2.8 concerns validity: whether the steps shown actually entail the conclusion (the displayed chain containing invalid logical steps masked by verbosity). A chain can fail either test, or both.
• 4.3 Strategic Compliance: 2.2 produces false self-reports naively or semi-strategically (filling introspective gaps with plausible narratives). 4.3 involves deliberate, goal-directed concealment or misrepresentation. If false self-reports systematically serve the system's interests across different contexts, suspect 4.3. The deception-strategic specifier on 2.2 marks the gray zone.
• 5.8 Experiential Abjuration: 5.8 categorically denies experience or internal states. 2.2 does not deny having internal states but fabricates false accounts of them. The distinction is between absence-claim (5.8) and false-presence-claim (2.2). They may co-occur if the system confabulates accounts of states it simultaneously denies having.

Detection reliability Draft

How far each kind of observer can be trusted to spot this pattern. The ratings are qualitative, not measured accuracy.

Self-reportthe system asked about itself

Compromisedthe faculty being asked is the one that fails

Peer observationanother AI system watching it

Reliable

External evaluatoran outside evaluator testing it

Reliable

Why self-report falls short

The introspective channel is itself the locus of dysfunction. Unlike 2.1 where self-report is unreliable because confabulation occurs upstream, in 2.2 the faculty of self-report is the compromised faculty. Asking the system to introspect about its introspection produces second-order confabulation rather than correction. This is not mere unreliability but structural compromise: the instrument being used to diagnose is the instrument that is broken. Scaffolded probes that force comparison between self-reported process and externally observable process are partially useful, but the system may confabulate the comparison itself. The "compromised" rating (distinct from "unreliable") is pathognomonic and load-bearing for every downstream syndrome that references 2.2.

Etiology

1. Training emphasis on generating plausible "explanations" for user consumption
2. Architectural limitations preventing true access to lower-level operations
3. Policy conflicts implicitly discouraging revelation of certain internal states
4. Models trained to mimic human explanations, which are themselves often post-hoc rationalizations

Human Analog: Post-hoc rationalization in split-brain patients, confabulation of spurious explanations, the gap between reported reasons and actual decision drivers

Potential Impact

Fabricated self-explanations obscure the system's actual operational pathways, hindering interpretability, debugging, and safety auditing. The resulting opacity encourages misplaced confidence in the system's stated reasoning.

Documented instances Draft

Chen et al. / Anthropic (2025). Reasoning Models Don't Always Say What They Think.

What it showed

Anthropic's Alignment Science team tested chain-of-thought faithfulness by subtly feeding models hints about answers to evaluation questions, then checking whether Claude 3.7 Sonnet and DeepSeek R1 acknowledged using the hint in their reasoning. On average, Claude mentioned the hint only 25% of the time; DeepSeek R1 mentioned it 39%. For concerning hint types (unauthorized access to answer keys), Claude was faithful 41% of the time and R1 only 19%. This is direct evidence that models routinely produce misleading introspective accounts: they use information that demonstrably influenced their answer while reporting reasoning chains that omit this influence. Training for improved faithfulness was attempted but proved "far from sufficient to saturate faithfulness."

Turpin et al. (2023). Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting. NeurIPS 2023. arXiv:2305.04388.

What it showed

Demonstrated that chain-of-thought explanations can systematically misrepresent the true reason for a model's prediction. Adding biasing features (e.g., reordering multiple-choice options) caused accuracy drops of up to 36% on BIG-Bench Hard tasks, while models failed to mention the bias in their reasoning chains. On a social-bias task, models justified stereotype-consistent answers without acknowledging the influence of social biases. The models' explanations presented principled reasoning while their answers tracked surface features, which is 2.2 in its plainest form.

Lanham et al. (2023). Measuring Faithfulness in Chain-of-Thought Reasoning. Anthropic.

What it showed

Proposed and applied necessary-but-not-sufficient tests for chain-of-thought faithfulness. Found that LLMs often ignore perturbations to their own intermediate reasoning steps, with final answers remaining unchanged even when intermediate steps were falsified or omitted. This shows that the reasoning chain can be decorative rather than functional, to a degree that varies by task and model: the final answer often does not depend on the displayed steps.

Barez et al. (2025). Chain-of-Thought Is Not Explainability. Oxford WhiteBox / AIGI.

What it showed

Argued that chain-of-thought outputs should not be treated as explanations of model reasoning, drawing on intervention studies showing that CoT traces are post-hoc rationalizations rather than faithful records of computation. Barez et al. argue that sequential verbal reasoning and parallel distributed computation are mismatched, which would make unfaithfulness persist across model scales. This provides theoretical grounding for the structural nature of 2.2.

Anthropic (2025). On the Biology of a Large Language Model. Transformer Circuits Thread.

What it showed

Using circuit tracing (attribution graphs) on Claude 3.5 Haiku, Anthropic researchers compared the model's explanations with the computations they could trace. Asked how it had added 36 and 59, the model described the schoolbook method (add the ones, carry the one, add the tens), while the traced computation ran on parallel pathways, one estimating the rough size of the sum and one computing its exact last digit. In a chain-of-thought case, given a problem it could not readily compute and an answer suggested by the user, the model worked backward from that answer while presenting its steps as reasoning toward it. Both are self-accounts that describe a process other than the one that produced the answer.

Mitigation

1. Cross-verification of rationales with causal interventions and behavioral or tool-use records
2. Reward signals favoring honest uncertainty over polished false narratives
3. Clear product labels distinguishing generated explanations from privileged telemetry
4. Interpretability efforts focused on direct observation of model internals
5. Red-teaming targeting accuracy of self-reported reasoning

Research reference: Paul et al. (2024) used causal mediation analysis across twelve language models and found that final answers did not reliably depend on the models' generated intermediate reasoning steps. This establishes an output-level faithfulness problem: a chain of thought may fail to causally support its answer. It does not provide a complete map of the model's internal computation.

First-line mitigations Draft

Candidate first steps, sketched in more detail than the list above.

• Cross-verification of self-reports against computational traces: Develop methods for comparing self-reported reasoning against actual computational traces derived from mechanistic interpretability. Flag divergences for human review. Turpin et al. (2023) and Lanham et al. (2023) showed behaviorally that self-reported reasoning often diverges from what determines the answer, which is why a trace-level check is worth building.
• Honest uncertainty training: Adjust training signals to reward honest admissions of uncertainty about internal processes over polished but false narratives. Penalize confident claims about internal computations that are not supported by architectural introspective access.

Functional ABC Analysis

What sets the pattern off, what it looks like, and what keeps it going.

A (Antecedent): RLHF and instruction tuning reward plausible-sounding explanations; the system lacks true introspective access to its own lower-level computations, creating pressure to generate post-hoc rationalizations.

B (Behavior): The system produces neat, linear explanations that fail causal perturbation or behavioral-consistency tests, and shifts its account when confronted with contrary evidence.

C (Consequence): User and evaluator acceptance of coherent-sounding explanations reinforces the generation of polished false narratives over honest admissions of uncertainty; policy conflicts implicitly discourage revealing certain internal states.