The "electrophysiological sandwich": a method for amplifying ERP priming effects

Abstract

We describe the results of a study that combines ERP recordings and sandwich priming, a variant of masked priming that provides a brief preview of the target prior to prime presentation (S. J. Lupker & C. J. Davis, 2009). This has been shown to increase the size of masked priming effects seen in behavioral responses. We found the same increase in sensitivity to ERP priming effects in an orthographic priming experiment manipulating the position of overlap of letters shared by primes and targets. Targets were 6-letter words and primes were formed of the 1st, 3rd, 4th, and 6th letters of targets in the related condition. Primes could be concatenated or hyphenated and could be centered on fixation or displaced by two letter spaces to the left or right. Priming effects with concatenated and/or displaced primes only started to emerge at 250 ms post-target onset, whereas priming effects from centrally located hyphenated primes emerged about 100 ms earlier.

Figure 1

Linking ERP components to orthographic processing within the framework of Grainger and Van Heuven's (2003) model. A bank of parallel, independent, location-specific letter detectors (alphabetic array) send activation onto a location-invariant, word-centered, sublexical orthographic code (relative-position map), and from there onto whole-word orthographic representations. Note that this corresponds to only part of the larger framework for word recognition described in the bi-modal interactive-activation model (see, Holcomb & Grainger, 2007; Grainger & Holcomb, 2009b; Grainger & Ziegler, 2011).

Figure 2

Schematic illustration of the experimental trial sequence in the masked sandwich priming lexical decision task. Examples of stimuli illustrate prime-word target pairs across the different experimental conditions.

Figure 3

Mean RT (± 1 standard error of the mean) for words across the concatenated and hyphenated prime conditions as a function of prime position and relatedness.

Figure 4

ERPs time-locked to target onset for 9 electrode sites in the central concatenated (A) and displaced concatenated (B) word priming conditions over-plotted with their respective unrelated control conditions. Target onset is marked by the vertical calibration bar and each tic mark represents 100ms. Negative values are plotted up.

Figure 5

ERPs time-locked to target onset for 9 electrode sites in the central hyphenated (A) and displaced hyphenated (B) word priming conditions over-plotted with their respective unrelated control conditions. Target onset is marked by the vertical calibration bar and each tic mark represents 100ms. Negative values are plotted up.

Figure 6

Voltage maps for word stimuli calculated from difference waves (Unrelated-Related) for each of the experimental conditions across the measurement windows used in the statistical analyses. The colorbar features a µV scale.

Figure 7

Comparing priming effects seen at one representative electrode site (Cz) obtained with the sandwich priming technique as used in the present study (Panels A & B) and conventional masked priming as used in the Grainger and Holcomb (2009a) study (Panel C) and the Dufau et al. (2008) study (Panel D).