The influence of word valence on the right visual field advantage in the VHF paradigm: time to adjust the expectations

ABSTRACT

Previous studies found that valence and visual half-field have an influence on word reading: Words are processed more efficiently when they evoke positive feelings and when they appear in the right visual field. In the present study we tried to address previous (contradictory) reports of an interaction between valence and visual half-field.A group of 39 right-handed undergraduates completed a lexical decision task in their native language (Dutch). They responded to 300 trials with real words and 300 trials with non-words. Overall, participants responded more efficiently to positive words and there was a strong right visual field advantage. We did not find a significant interaction, however. Further analysis indicated that to find a replicable interaction between a stimulus characteristic and visual half-field, one requires much high numbers of participants and stimuli than done so far. Experimental power is particularly low when the interaction is not fully crossed (a right visual field advantage for one type of stimulus and an equally large left visual field advantage for the other type of stimulus). If such investment cannot be made, the outcome is likely to be ambiguous at best and deceiving at worst if only significant findings are published.

KEYWORDS:

• Visual half field
• valence
• right visual field advantage
• laterality

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The files with the raw data and the R file used for analysis are available at https://osf.io/jzqac/.

Notes

1 Kuruoglu and Mergen (2016) did not report performance for neutral stimuli but the positive words were responded to more rapidly than the negative words. Mohr et al. (2005, 2008) did not contain all conditions.

2 See Vasilev, Yates, and Slattery (2019) for a similar conclusion in a related domain.

3 For comparison, our study included 39*150*2*2 = 23,400 test observations, of which half could not be used because they were responses to nonwords.