In this paper the relation between subjective difficulty, features of stimuli and improvement of efficiency is tested in a visual search experiment.
The relation is tested, in hope of getting a better understanding of the relation between perception and subjective difficulty.
11 subjects are tested in a web-based hybrid foraging search experiment, where each subject conducts a visual search for three different targets per trial.
There is a total of 20 trials where targets and distractors in each trial are randomised in size, difficulty condition and color.
At the end of each trial the subjects rates the subjective difficulty of the search.
A manipulation check is conducted with a linear mixed model for the response time and setsize typical for visual search.
The manipulation check shows the linear relation and that the search efficiency is dependent on visual features, which is in accordance with visual search models.
The dependence between subjective difficulty and the perception of visual features is tested with a linked linear mixed model, using the predictions of the model for response time from the manipulation check.
The model shows a relation between the subjective difficulty, the visual features, the predictions and residuals from the manipulation check model.
Lastly the interaction between trial and subjective difficulty is used in a linear mixed model to predict efficiency improvements for the visual search.
The model shows a significant predictability by the interaction between subjective difficulty and trial for response time in the visual search paradigm, with results indicating a growing subjective difficulty will hinder improvement.
The results of the findings are discussed for implications on user centered product design, further research and the relation between subjective difficulty and visual search.