Emotion improves and impairs early vision

BR Bocanegra, Rene Zeelenberg

Research output: Contribution to journalArticleAcademicpeer-review

145 Citations (Scopus)


Recent studies indicate that emotion enhances early vision, but the generality of this finding remains unknown. Do the benefits of emotion extend to all basic aspects of vision, or are they limited in scope? Our results show that the brief presentation of a fearful face, compared with a neutral face, enhances sensitivity for the orientation of subsequently presented low-spatial-frequency stimuli, but diminishes orientation sensitivity for high-spatial-frequency stimuli. This is the first demonstration that emotion not only improves but also impairs low-level vision. The selective low-spatial-frequency benefits are consistent with the idea that emotion enhances magnocellular processing. Additionally, we suggest that the high-spatial-frequency deficits are due to inhibitory interactions between magnocellular and parvocellular pathways. Our results suggest an emotion-induced trade-off in visual processing, rather than a general improvement. This trade-off may benefit perceptual dimensions that are relevant for survival at the expense of those that are less relevant.

Results from a variety of paradigms indicate that emotion affects perceptual processing. For example, evidence for enhanced identification of emotionally significant stimuli has been obtained for briefly presented and masked arousing words (Anderson & Phelps, 2001; Zeelenberg, Wagenmakers, & Rotteveel, 2006) and for emotional faces, snakes, and spiders in visual search tasks (Fox et al., 2000; Öhman, Flykt, & Esteves, 2001). Also, brain-damaged patients who exhibit spatial neglect show less visual extinction for faces with happy or angry expressions than for faces with neutral expressions (Vuilleumier & Schwartz, 2001).

The effects of emotion on perception could arise at different stages of visual processing (Vuilleumier, 2005). That is, emotion-induced effects may be mediated by specialized neural systems dedicated to the recognition of certain classes of stimuli, such as faces, words, and objects. Recent studies, however, indicate that even earlier stages of visual processing are affected by emotion. Evidence that emotion enhances early perceptual processing comes primarily from research in cognitive neuroscience (for a review, see Vuilleumier, 2005). Neuroimaging studies using positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) have shown that emotional images, compared with neutral images, enhance activations in early occipital cortex (Lang et al., 1998; Morris et al., 1998). Additionally, event-related potential (ERP) studies (Schupp, Junghöfer, Weike, & Hamm, 2003) suggest that early visual components are affected by the emotional significance of stimuli. Some studies have indicated that modulations of ERPs may be observed as early as 60 to 90 ms after stimulus onset (Stolarova, Keil, & Moratti, 2006).

Although the modulation of early visual processing by emotion has now been demonstrated in a number of studies, the behavioral consequences of this modulation remain largely unknown. The first and only behavioral study showing that emotion enhances low-level vision was published only recently. In this study, Phelps, Ling, and Carrasco (2006) presented fearful or neutral face cues prior to a target Gabor patch presented at low luminance contrast. Results showed that the fearful face cues enhanced threshold contrast sensitivity. The mechanisms underlying emotion-induced enhancement of low-level vision, however, are still largely unspecified. As a consequence, it is not clear whether all early dimensions of vision are enhanced by emotion, or whether the effects are restricted to certain basic visual attributes.

It has been proposed that the amygdala, a medial temporal lobe structure involved in emotional processing, may boost sensory processing via connections to the visual cortex (Anderson & Phelps, 2001; Morris et al., 1998). That is, emotional stimuli activate the amygdala (Vuilleumier, 2005), which in turn may modulate ongoing processing in the visual cortex. Neuro-anatomical studies with primates indicate that the amygdala projects to the earliest levels of the ventral visual stream (Amaral, Behniea, & Kelly, 2003). A common distinction within the visual system is that between magnocellular and parvocellular visual channels. These channels differ in the nature of the visual information they carry. Whereas magnocellular channels process predominantly coarse, low-spatial-frequency (LSF) information, parvocellular channels process predominantly fine-grained, high-spatial-frequency (HSF) information. It has been shown that projections from the amygdala to occipital cortex are mostly of the magnocellular type, and therefore carry predominantly LSF information (Amaral et al., 2003). This raises the following question: Are the facilitatory effects of emotion restricted to the processing of LSF information, or is HSF information also affected? The fact that projections from the amygdala to early visual cortices are magnocellular (Amaral et al., 2003) suggests that only LSF information may benefit from the presentation of a fearful face.
Original languageEnglish
Pages (from-to)707-713
Number of pages7
JournalPsychological Science
Issue number6
Publication statusPublished - 1 Jun 2009


Dive into the research topics of 'Emotion improves and impairs early vision'. Together they form a unique fingerprint.

Cite this