Coded turns from 429 human-AI pair programming sessions across 34
projects, in long format: human_long holds the human turns (10,796
rows), ai_long the AI turns (8,551 rows). Each session's ordered
codes form one categorical sequence, which makes the pair a natural
two-cohort input for the higher-order verbs – e.g.
hg_bootstrap(human_long, action = "code", actor = "session_id", time = "timestamp") or hg_compare(human_long, ai_long, ...).
Format
Data frames in long format with 9 columns:
- message_id
Integer. Turn index.
- project
Character. Project identifier (Project_1 .. Project_34).
- session_id
Character. Unique session hash.
- timestamp
Integer. Unix timestamp for ordering.
- session_date
Character. Date of the session (YYYY-MM-DD).
- code
Character. Interaction code.
- cluster
Character. High-level cluster: Directive, Evaluative, or Metacognitive (human codes); AI turns carry their own scheme.
- code_order
Integer. Order of the code within the session.
- order_in_session
Integer. Absolute turn order within the session.
An object of class data.frame with 10796 rows and 9 columns.
An object of class data.frame with 8551 rows and 9 columns.
Source
Saqr, M. (2026). Human-AI vibe coding interaction study. https://saqr.me/blog/2026/human-ai-interaction-cograph/
Details
The same data feed all three structure families: the ordered codes are
sequences for the memory family, a session is a natural hyperedge over
the codes that co-occur in it (group_hypergraph()), and a fitted
memory network becomes a pathway complex for the simplicial family
(simplicial() with type = "pathway").
Examples
bs <- hg_bootstrap(human_long, action = "code", actor = "session_id",
time = "timestamp", n_boot = 20, max_order = 2,
seed = 1)
rules <- hg_get(bs, order_min = 2)
head(rules)
#> from to order count probability ci_lower ci_upper
#> 1 Frustrate -> Specify Command 2 6 0.05405405 0.02738722 0.07904095
#> 2 Frustrate -> Specify Correct 2 9 0.08108108 0.03276329 0.12788847
#> 3 Frustrate -> Specify Frustrate 2 21 0.18918919 0.11566507 0.24824375
#> 4 Frustrate -> Specify Inquire 2 8 0.07207207 0.03930894 0.10455508
#> 5 Frustrate -> Specify Interrupt 2 9 0.08108108 0.03793060 0.11064680
#> 6 Frustrate -> Specify Refine 2 7 0.06306306 0.02103135 0.10300585
#> support n_boot_used
#> 1 0.95 20
#> 2 0.95 20
#> 3 0.95 20
#> 4 0.95 20
#> 5 0.95 20
#> 6 0.95 20