Decision Sequence Timeline¶
Analysis artifact — instantiates Cascade Initiation Bias Diagnosis and Correction
A timeline showing when visible actors acted and when later actors could observe those actions.
A cascade always arrives as a story — "everyone was moving," "the market already knew," "the whole team was on board." Decision Sequence Timeline replaces that story with a strictly ordered event log annotated with visibility edges: for each action it records not only when it happened but which earlier actions the actor could actually see at the moment they decided. The one move that makes it this mechanism and not a sibling is that it treats observability as data. It never asks why anyone acted — only who acted, in what order, and who could have been watching whom. That skeleton is what lets a later analyst tell a chain of independent decisions apart from a chain of people all watching the same falling tape.
Example¶
A small-cap stock drops 18% in a morning and the desk narrative is that "sellers know something." The timeline is built from the trade tape. Ordered by timestamp, it shows a single large block sale at 9:41 — later confirmed to be a fund meeting a redemption, not acting on news — followed by a steepening run of smaller sells, each placed after the price ticked visibly lower. A visibility edge is drawn from the falling quote to every subsequent seller: they could all see the drop; none could see the reason. The mechanism marks 9:41's block and the two mid-morning momentum funds as threshold actors — the points where the selling rate jumped. The output is one artifact: a sequence in which one liquidity-driven trade is observed and echoed by many, not many independent judgments of bad news. That structure is what the rest of the correction hangs on.
How it works¶
- Assemble a single ordered event log — every visible action (trade, endorsement, exit, vote, post) with actor, timestamp, and the action taken.
- Draw visibility edges — for each event, record which prior events were observable to that actor when they decided. This is the step that separates a timeline from a mere list.
- Mark threshold actors — flag the inflection points where the joining rate accelerated, distinguishing chronological first movers from later actors whose participation tipped the slope.
- Bound the upstream edge — decide where the reconstruction stops, so the artifact stays finite without amputating the true origin.
Tuning parameters¶
- Time resolution — millisecond tape versus day-level adoption log. Finer resolution resolves who-saw-whom but costs collection effort and can invent order where none is meaningful.
- Visibility model — what counts as "observable" (a public tape is fully visible; a private endorsement is not). A generous model over-connects the graph; a strict one may miss real observation channels.
- Actor granularity — individuals, desks, or firms. Coarser nodes are cheaper but can hide an internal cascade inside a single "actor."
- Threshold definition — how large a jump in joining rate marks an accelerator. Set it loose and everyone looks pivotal; set it tight and you miss the real tipping actors.
- Inclusion boundary — how far back and how wide to trace before declaring the sequence's origin.
When it helps, and when it misleads¶
Its strength is that it converts an impression of momentum into an inspectable order-and-visibility structure, and it makes one thing legible that nothing else does: whether later actions could even have carried independent information, or whether each one merely observed the ones before it.
Its failure mode is baked into its scope: the timeline shows sequence, not content. It can prove that B acted after and in view of A, but never that B acted because of A — post hoc is not propter hoc. The classic misuse is reading temporal order as proof of imitation when the later actors in fact held their own private signals and would have acted regardless. This is exactly the trap an informational cascade sets — rational actors can rationally follow predecessors, so order alone underdetermines the mechanism.[1] The guarding discipline is to treat the timeline as a map of where to look, never a verdict: hand its threshold actors and visibility edges to source-recovery mechanisms before concluding that a sequence was hollow.
How it implements the components¶
cascade_adoption_chain_map— the artifact is this map: the ordered, visibility-annotated sequence of adoptions, distinguishing first movers from followers.threshold_actor_marker— it flags the actors at the acceleration inflection, so later diagnosis knows which nodes to interrogate first.
It does not recover what any initiator actually knew — that belongs to Initiator Interview Protocol via initiator_information_source_record — nor trace repeated claims back to a common origin, which is Common Source Citation Check's source_chain_reconstruction.
Related¶
- Instantiates: Cascade Initiation Bias Diagnosis and Correction — supplies the ordered sequence-and-visibility scaffold the rest of the diagnosis reasons over.
- Sibling mechanisms: Initiator Interview Protocol · Evidence Provenance Checklist · Private Signal Survey · Blind Independent Vote Reset · Source Disclosure Brief · Network Position Review · Common Source Citation Check
Editorial Notes¶
Form Classification¶
Form family: Record, Log & Register
Rationale: The timeline preserves actual actor actions, timestamps, visibility edges, threshold actors, and the bounded upstream history of a decision sequence, so its operative form is an event and information-provenance record.
Nearest alternative: Representation, Specification & Plan — The timeline visually externalizes sequence, but its value depends on recording what actors actually did and could observe rather than a prospective process map.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Behavioral Economics
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Behavioral information economics cohered informational-cascade models in which observable predecessor actions can rationally swamp later actors' private signals.
Related originating lineages:
- Economics & Finance — Sequential social-learning models supplied formal event order and information-set reconstruction.
- Sociology & Anthropology — Diffusion and conformity research supplied empirical analysis of visible adoption sequences.
Review resolution: Behavioral information economics cohered informational-cascade models in which observable predecessor actions can rationally swamp later actors' private signals.
Attribution caveat: The causal theory is established; annotating a timeline with actor-specific visibility edges is a synthesized diagnostic artifact.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
References¶
[1] Informational cascade — Bikhchandani, Hirshleifer & Welch (1992): once the public sequence of predecessors' actions outweighs an actor's own private signal, it can be individually rational to ignore that signal and follow the crowd, so a long chain of "agreement" may encode almost no independent information. withdrawn registry ↩