Slider rating trial

Rate each stimulus on a configurable slider scale inside an experiment trial.

When to use it

Use this when each trial needs a continuous or fine-grained judgement of a stimulus: valence, arousal, confidence, attractiveness, perceived duration. Discrete buttons force the judgement into a handful of categories and invite response sets, while a slider records a value on the scale you define. If the ratings belong on survey pages rather than in a timed experiment flow, use a form trial with the slider question type instead; if you only need a few labelled categories, buttons are simpler.

Trial file

type	stimFormat	stim1	responseType	responseOptions	required
test	.jpg	scene_forest	slider	0_100_5;50	1
test	.jpg	scene_beach	slider	0_100_5;50	1
test	.jpg	scene_market	slider	0_100_5;50	1
test	.jpg	scene_highway	slider	0_100_5;50	1
test	.jpg	scene_garden	slider	0_100_5;50	1

How it works

responseType set to slider replaces the default response mode with a horizontal slider under the stimulus. Left unconfigured, that slider runs 0 to 100 in steps of 1; the responseOptions cell is where you change that. Its first segment uses the min_max_step grammar: 0_100_5 gives a 0 to 100 scale in steps of 5. The next semicolon-separated segment is the start position, here 50, so the handle begins at the midpoint rather than at the minimum default.

required matters more than it looks. A slider trial normally shows a Next button immediately, and a participant who clicks straight through records the start value as their rating, indistinguishable from a deliberate 50. Any non-empty value in required hides that button until the slider has been touched, so every saved rating reflects an actual movement or click on the scale.

Keyboard responses are disabled while responseType is slider, so there is no risk of a stray keypress ending the trial; the participant sets the handle and confirms with the button.

One spelling warning: the slider flags in responseOptions are case-sensitive. hideStart, hideValue, hideGrid, hideMinMax and removeFill must be written exactly like that, and a misspelt flag is silently ignored.

Variations

• Label the endpoints by writing them in parentheses in responseOptions, for example 0(not at all)_100(extremely);50.
• Swap the numeric scale for a discrete labelled one: poor_average_good;average in responseOptions makes a three-position slider recorded as values 1 to 3, starting at the middle label.
• Start with no visible handle by adding hideStart in responseOptions; an untouched slider then records NA instead of a start value, which is the honest alternative to forcing a touch with required.
• Reduce anchoring cues with hideValue (no live numeric readout) or removeFill (no colour fill up to the handle) in responseOptions.