Skip to content

Diagnostic notices

Purpose

MonStim shows notices beside the selected session, dataset, or experiment when it finds a condition worth reviewing. A warning means that the condition can prevent a valid analysis or materially affect it. An information notice is a useful check, not a diagnosis and not an automatic reason to discard data.

Read the notice text itself: it identifies the affected level and usually the window, channel, or recording involved. Notices update after relevant edits and after undo or redo.

Diagnostic symbols

The selection pane shows one symbol beside the selected experiment, dataset, or session when it has notices. The symbol summarizes the highest severity present; open its tooltip to read every notice and its details.

Symbol Meaning What to do
⚠ Warning. At least one notice could prevent a valid analysis or materially affect it. Warnings take precedence when warnings and information notices occur together. Review and resolve the affected condition before making a final plot or export whenever possible.
ℹ Information. The selected item has useful review information but no warnings. Read the tooltip and confirm that the condition is expected for the analysis.

Steps: what to do first

  1. Resolve warnings before creating a final plot or export whenever possible.
  2. Inspect the affected session before changing a dataset or experiment.
  3. Keep an intentional exception only when it is justified in the analysis record.
  4. Do not change windows simply to silence an overlap notice; overlapping M- and H-response windows can be scientifically appropriate. Review the trace and method instead.

Session notices

Notice What it means Suggested check
No active recordings Every recording is excluded, or none is available. Review recording exclusions and restore recordings if appropriate.
Non-positive window duration A latency window has zero or negative duration. Correct the duration before analysis.
Window outside acquisition bounds A window begins before the available signal or ends after it. Check the recording duration and the window start/duration.
Windows overlap substantially Two windows overlap by at least half of the shorter window on a channel. Confirm that overlap is intended; see Extrema peak-to-trough methods if using an extrema method.
Missing M-wave window No window has a recognized M-wave name. Add or rename the M-wave window if M-max normalization is intended.
Recording consistency warning Recordings differ in sampling rate, channel count, stimulus delay, or stimulus ordering. Verify import files and avoid treating incompatible recordings as interchangeable.

Dataset and experiment notices

Notice What it means Suggested check
Different latency-window sets Child sessions or datasets do not use the same named, ordered windows. Inspect the affected sessions. Standardize windows only if a shared analysis definition is intended.
Mixed sampling rates Child sessions or datasets have different sampling rates. Confirm that the comparison remains appropriate after filtering and windowing.
No active sessions or datasets Nothing remains included at that level. Review session or dataset exclusions.
One active session A dataset has only one active session. Interpret any aggregate result as a single-session result, not a replicate estimate.
High diversity of window names Many window names occur only rarely within a dataset. Check for spelling variants and unintended exploratory windows before aggregating.

M-wave naming

For M-max-related checks, MonStim recognizes the global M-wave Recognition Names without regard to case. The shipped list is M-wave, M_wave, M wave, Mwave, M-response, M_response, and M response, and you can replace it in File > Settings Center > Global Analysis > Latency windows. An empty list disables automatic M-wave recognition. Use a consistent name across comparable sessions to make plots and notices easier to interpret. See M-max estimation and review for the complete setup guidance.