Mechanism
Inquiry mode is the structured cognitive process the confidence governor activates when the agent's confidence falls below the execution authorization threshold but remains above a minimum engagement threshold. It is disclosed in Chapter 5 of the cognition filing as a "pause-to-think" mode: the agent stops executing and redirects its cognitive resources toward information acquisition, hypothesis evaluation, and uncertainty resolution. Inquiry mode is not a passive waiting state. It is an active cognitive process in which the agent systematically identifies the factors contributing to low confidence and generates targeted inquiry operations designed to address those factors.
The mode exists because of a prior architectural commitment: execution and cognition are structurally separated. When the confidence governor withdraws execution authorization, it gates only the execution pathway. The cognitive pathway remains active. An agent under confidence suspension therefore enters a non-executing cognitive mode in which it is fully cognitively active but structurally prohibited from acting. Inquiry mode is what the agent does with that preserved cognitive capacity: rather than idling until external conditions change, it works to resolve the conditions that drove confidence down.
The Engagement Band
Inquiry mode occupies a band, not a point. The confidence governor recognizes a gating threshold below which execution is no longer authorized, and a separate minimum engagement threshold below which inquiry itself is no longer warranted. Inquiry mode is activated specifically when confidence has dropped below the execution authorization threshold but has not dropped below the minimum engagement threshold. This places inquiry mode between ordinary authorized execution above and disengagement below.
The distinction matters because it separates two different conditions. An agent whose confidence is merely insufficient to act may still be sufficient to reason productively about why it cannot act and what would change that. An agent whose confidence has collapsed past the engagement threshold is in a different regime, one the chapter routes through suspension, waiting states, or, in severe cases, the locked state reserved for governance-mandated halts. Inquiry mode is the response to the recoverable middle.
The Structured Inquiry Operations
The chapter enumerates inquiry mode as a plurality of structured inquiry operations. The first is hypothesis expansion. During normal execution the agent's hypothesis generation is constrained by execution urgency: it focuses on the most probable hypotheses to minimize deliberation time. During inquiry mode the agent is freed from that urgency and generates and evaluates less probable but potentially valuable hypotheses that would have been pruned during normal execution. The agent broadens its set of hypotheses about the current task state, the environmental conditions, and the available strategies.
The second operation is information ingestion. The agent identifies specific information gaps that are contributing to uncertainty and generates requests for that information. The requests may be directed externally, to databases, APIs, human supervisors, or collaborating agents, or directed internally, prompting the agent to re-examine its own memory and experiential records for relevant information that was not surfaced during initial task assessment.
The third operation is re-evaluation loops. The agent re-examines previously made assessments, decisions, and intermediate results in light of the conditions that caused confidence to drop. A decision that appeared sound under higher confidence may reveal weaknesses when reconsidered under the scrutiny of inquiry mode. Re-evaluation lets the agent detect and correct errors that would otherwise propagate through subsequent execution.
The fourth operation is condition monitoring. The agent monitors the conditions that caused confidence to drop and evaluates whether they are improving, stable, or worsening. If the adverse conditions are transient, such as a temporary resource shortage or a brief communication interruption, the agent may determine that waiting for conditions to improve is the appropriate strategy. If the conditions are chronic or worsening, the agent may determine that a more active response is required, such as redirecting to an alternative strategy or escalating to a human supervisor.
The Iterative Loop
Inquiry mode operates iteratively. At each iteration the agent evaluates the results of its most recent inquiry operations, updates its confidence computation with the new information or revised assessments, and determines whether confidence has recovered sufficiently to warrant a transition back to execution authorization. If confidence has recovered, the agent exits inquiry mode and resumes execution through the recovery pathway. If confidence has not recovered, the agent generates a new set of inquiry operations and continues the cycle.
Recovery from inquiry mode is not a bare threshold crossing. The chapter's recovery pathway requires confidence restoration, then a stability verification phase, then reauthorization. During stability verification the confidence governor monitors the confidence value and its trajectory over a verification period to confirm that the restored confidence is stable and not merely a transient spike, and the confidence value must exceed the authorization threshold by a hysteresis margin throughout. Only after stability is verified does the governor reconnect the execution pathway. Information gathered and hypotheses generated during inquiry are carried forward into the resumed execution plan rather than discarded.
Curiosity as an Affective Modulator
The agent's affective state field includes a curiosity dimension that modulates how the agent responds to confidence interruption. Curiosity biases the agent toward inquiry and exploration rather than disengagement and passivity when confidence drops below the execution authorization threshold. An agent with elevated curiosity responds to interruption with an intensified inquiry response: it generates more hypotheses, seeks more information, explores a wider solution space, and persists in inquiry mode longer before disengaging. An agent with suppressed curiosity responds conservatively, performing minimal inquiry, preserving state, and waiting for external conditions to change.
The chapter describes two curiosity orientations that produce different inquiry behaviors. An internal curiosity orientation directs inquiry inward, toward the agent's own memory, reasoning processes, prior assessments, and affective responses, producing introspective inquiry that may reveal agent-side factors behind the confidence deficit. An external curiosity orientation directs inquiry outward, toward the environment, the task domain, available information sources, and collaborating agents, producing expansive inquiry that may discover task-side or environment-side factors relevant to recovery. Curiosity does not override the gate. An agent with high curiosity that is in a suspended execution state remains suspended; curiosity modulates the quality and intensity of the non-executing cognitive activity, not the gating decision itself.
Coupling to Discovery Traversal
Inquiry mode couples to discovery traversal of the semantic index. When the agent is navigating the index through a sequence of anchor-to-anchor transitions and its confidence drops below the execution authorization threshold during traversal, the traversal pauses at the current anchor node. The agent does not advance to a new anchor. Instead it redirects its cognitive resources toward deeper evaluation of the current anchor's semantic neighborhood, in a structured operation the chapter calls an anchor neighborhood inquiry.
In the anchor neighborhood inquiry the agent re-evaluates the candidate transitions available from the current anchor, applying the expanded hypothesis generation and re-evaluation of inquiry mode; explores the anchor's semantic neighborhood in greater depth than normal traversal, examining related containers, edge-case connections, and less probable transition candidates that would be pruned under normal traversal urgency; and generates internal queries about the structure of the semantic space surrounding the anchor. The effect is that low confidence during discovery does not cause the agent to advance blindly into unfamiliar semantic territory. It instead deepens the agent's understanding of the territory it currently occupies, building a more robust basis for advancement once confidence recovers. An agent with elevated curiosity explores the neighborhood more broadly and persistently, sometimes discovering connections that simultaneously resolve the confidence deficit and advance the discovery objective.
From Inquiry to Waiting
Inquiry mode can resolve into a waiting state. The chapter defines a waiting state as a suspension sub-state the agent enters once it has completed its initial inquiry and assessment, determined that no productive cognitive action is available in the immediate term, and elected to defer re-evaluation until a specified trigger. Waiting-state triggers may be temporal, re-evaluate after a duration, or conditional, re-evaluate when an environmental condition is met, a resource becomes available, or a collaborating agent reports a confidence change.
The waiting state is not idle. The agent continues to monitor a reduced set of critical conditions, including catastrophic failure indicators, governance-mandated interrupts, and the waiting-state trigger conditions, and responds immediately if any of these change. Temporal reauthorization is not automatic: the passage of time alone does not restore execution authorization. At the trigger point the confidence governor performs a full confidence re-evaluation incorporating any changes that occurred during the wait. If the re-evaluation produces sufficient confidence above the threshold and its hysteresis margin, execution is reauthorized; otherwise the agent may enter a new inquiry cycle, schedule a new deferred evaluation, or escalate to governance infrastructure.
Why It Is Structured Rather Than Reactive
The chapter frames the confidence governor as proactive where conventional systems are reactive. Conventional autonomous agent runtimes that provide pause and resume capabilities suspend execution in response to external failures or resource interruptions. The confidence governor suspends execution based on the agent's own continuously computed assessment of its sufficiency, enabling the agent to stop itself before damage occurs rather than recovering after damage has occurred. Inquiry mode is the constructive face of that proactive stop: a pause that is spent reasoning about the deficit rather than blocked on it.
What distinguishes inquiry mode is that it is structurally integrated rather than bolted on. It is gated by the same confidence governor that gates execution, it preserves the full cognitive pathway through the structural separation of execution from cognition, its operations are enumerated and recorded, and its exit is governed by the same stability-verified recovery pathway that governs any return from suspension. The agent that pauses to think is not a degraded agent producing a best guess. It is a fully cognitively active agent that is prohibited from acting and is using that prohibition to make itself ready to act.
Disclosure Scope
Inquiry mode, comprising the confidence governor's activation of a structured pause-to-think process when confidence falls below the execution authorization threshold but remains above a minimum engagement threshold, the structured inquiry operations of hypothesis expansion, information ingestion, re-evaluation loops, and condition monitoring, the iterative evaluation cycle that updates confidence and either resumes execution through the stability-verified recovery pathway or generates a further inquiry cycle, the curiosity modulation of inquiry intensity and its internal and external orientations, the coupling to discovery traversal through anchor neighborhood inquiry, and the resolution into deferred waiting states with temporal and conditional reauthorization, is disclosed in the cognition filing (U.S. Application No. 19/647,395 and its international counterpart) in Chapter 5, principally at Section 5.8. This article describes that disclosed mechanism. The scope extends to embodiments in which the inquiry operations are realized over different external and internal information sources and over different index, anchor, and neighborhood representations, provided the inquiry process remains gated by the confidence governor, preserves the cognitive pathway under execution suspension, and returns to execution only through stability-verified reauthorization.
