International Affective Picture System Images: A Practical Researcher Guide

International Affective Picture System images, usually called IAPS images, are standardized color photographs developed for experimental research on emotion and attention. The set was created so researchers could present visual stimuli with published normative ratings instead of selecting emotionally evocative pictures only by intuition. IAPS materials are associated with the Center for the Study of Emotion and Attention at the University of Florida and with the work of Peter J. Lang, Margaret M. Bradley, and Bruce N. Cuthbert. Researchers commonly use the images to manipulate or measure affective responses in studies of attention, memory, psychophysiology, neuroscience, clinical psychology, and related fields. The best-known normative dimensions are valence, arousal, and dominance. Importantly, an IAPS image is not simply a “positive,” “negative,” or “neutral” photograph. Each stimulus has an identifier and normative data that allow researchers to select pictures for a defined region of affective space. Because some images depict disturbing, threatening, sexual, medical, or violent content, investigators should also consider participant safety, ethics approval, informed consent language, and debriefing. Researchers should obtain materials through authorized channels and follow the current terms of use rather than copying images from papers or websites.

Do not assume that IAPS images are an unrestricted public image library. The materials have historically been distributed for research through the University of Florida’s Center for the Study of Emotion and Attention, and research articles commonly direct investigators to request authorized access. Distribution arrangements can change, so the safest approach is to check the current CSEA or University of Florida instructions before planning a study. Avoid downloading collections from unofficial mirrors, file-sharing sites, course folders, or screenshots embedded in published papers. Those copies may be incomplete, mislabeled, altered, or distributed without permission. Unauthorized redistribution can also undermine the scientific value of a standardized stimulus set because widespread public familiarity may change participants’ reactions. If access to IAPS is unavailable or unsuitable, consider a documented alternative such as OASIS, NAPS, or GAPED, then verify its license and normative population. Your methods section should state exactly which stimulus database, version or citation, image identifiers, selection criteria, and rating dimensions were used. For publication, reproduce actual stimulus images only when the relevant rights and journal requirements allow it.

Valence describes how pleasant or unpleasant a stimulus is experienced, arousal describes the intensity or activation associated with the response, and dominance describes the perceived sense of control versus being controlled or overwhelmed by the emotional state. In IAPS research, these dimensions are typically treated as continuous rather than as simple labels. This matters because two pictures can have similarly negative valence but very different arousal, or similarly high arousal but opposite valence. Researchers often select stimuli by matching one dimension while manipulating another. For example, a study may compare pleasant and unpleasant high-arousal pictures while keeping average arousal as similar as possible across conditions. Another study may contrast unpleasant high-arousal images with neutral low-arousal images, but then the effects cannot be attributed to valence alone because arousal also differs. The normative scores are group averages, not guarantees about every participant. Age, culture, sex, clinical characteristics, familiarity, and experimental context can influence responses. Therefore, investigators should report the normative source used and consider whether local validation or manipulation checks are needed for their sample.

Start with the hypothesis, not with the pictures. Define which affective dimension or content characteristic must differ between conditions and which characteristics should be matched. Then use the normative ratings to create candidate pools. If valence is the independent variable, try to avoid unintentionally creating large arousal differences unless that is part of the design. If arousal is the target, match valence as closely as practical. Next, review semantic content because picture categories themselves can become confounds: threat, injury, food, romance, animals, faces, and landscapes may differ in more than emotional ratings. Also consider luminance, visual complexity, presence of people, social relevance, and image familiarity when these features could affect the dependent measure. Pre-register selection rules when possible, document image identifiers, and keep a reproducible stimulus-selection file. For culturally distinct or age-specific samples, check whether published local norms exist or pilot the stimuli. Finally, confirm that the planned exposure is ethically appropriate for your participants and that your consent, exclusion criteria, stopping procedures, and debriefing address potentially distressing content.

Do not assume that research access automatically includes permission to republish the photographs. Access for experimental use and permission to reproduce an image in a thesis, article, poster, presentation, website, or supplementary file can be different rights. Before reproducing IAPS pictures, check the current terms supplied with the materials, your institution’s guidance, and the publisher or university requirements. Where reproduction is not clearly permitted, a safer scholarly practice is to report IAPS image identification numbers, normative ratings, selection criteria, and the standard technical-report citation without embedding the actual photographs. This also helps other qualified researchers reproduce your stimulus selection while avoiding unnecessary redistribution. If a journal asks for representative stimuli, obtain permission where required or use a schematic/placeholder that does not reproduce protected content. Captions and methods should never imply that Contentxprtz, your institution, or the publisher owns the IAPS images. When editing a manuscript, keep a permissions log for every third-party figure, image, test, and stimulus set so rights issues can be resolved before final submission.

IAPS has been used internationally, but researchers should not assume that normative ratings are identical across populations. A 2023 systematic review found extensive cross-national use and validation while also emphasizing cross-cultural, age, and sex differences. Individual studies have reported differences between populations in valence or arousal for subsets of pictures. This is expected because affective meaning depends partly on experience, cultural context, age, and social norms. If your sample differs substantially from the original normative population, search for published norms from a comparable population and report which norms you used. When no suitable norms exist, a pilot rating study can help determine whether your selected stimuli produce the intended ordering in your target group. Do not “correct” original IAPS scores silently. Keep the original normative values and local pilot values separate, explain the rating scale and sample, and specify whether local ratings were used for stimulus selection, manipulation checks, or exploratory analysis. This transparency improves interpretability and makes cross-cultural findings easier to evaluate.

The main ethical issue is that some affective stimuli can be disturbing. Depending on the subset, pictures may include injury, threat, violence, medical scenes, sexual content, or other material that participants may find upsetting. The correct safeguards depend on the study population and intensity of exposure. Researchers should describe the stimulus categories accurately in the ethics application, use appropriate inclusion and exclusion criteria, provide informed consent language that does not compromise the hypothesis unnecessarily, and create a clear procedure for withdrawal or stopping the task. Vulnerable populations may require additional review. The protocol should also address privacy and data security when physiological or neuroimaging data are collected. During analysis and publication, use stimulus identifiers rather than redistributing images unless permission allows reproduction. A debriefing should explain the purpose of emotional stimulation and provide appropriate support information when warranted. These considerations do not make IAPS inappropriate; they make careful protocol design essential. Your institutional review board or ethics committee remains the authority for the specific study.

The Self-Assessment Manikin, commonly abbreviated SAM, is a nonverbal pictorial rating method frequently used in affective research to capture dimensions such as valence, arousal, and dominance. Its visual format reduces reliance on long verbal labels and has been used in the normative work associated with affective stimulus sets. In a study, participants may view a picture and then choose the SAM figure that best represents how pleasant versus unpleasant, activated versus calm, or dominant versus controlled they feel. Researchers should follow the scoring procedure specified by the source they are using and report the scale direction clearly, because readers must know what higher and lower numbers mean. If you modify the response format, number of scale points, display timing, or instructions, describe those changes instead of claiming an exact replication of the original protocol. For multilingual or cross-cultural studies, SAM can reduce some language burden, but it does not eliminate cultural differences in emotional response. Pilot testing remains useful when the target population or experimental context differs substantially from the normative sample.

Several standardized affective image sets may be suitable, depending on your design. OASIS was introduced as an open affective image set with 900 images and normative valence and arousal ratings, and its authors specifically highlighted fewer copyright restrictions for research use. NAPS provides a large set of realistic high-quality photographs with affective ratings and technical image properties; its research-access conditions should be checked before use. GAPED was developed to expand the availability of affective pictures, including specific negative categories plus positive and neutral images, and the Swiss Center for Affective Sciences provides research-material information. No alternative is automatically interchangeable with IAPS. Compare the rating dimensions, scale range, normative sample, content categories, image resolution, physical properties, licensing terms, and evidence of validation. If your preregistration or prior literature specifies IAPS, switching databases can affect comparability, so explain the reason and update the design transparently. A well-documented alternative is better than an unofficial or incomplete copy of IAPS.

Report enough information for a qualified researcher to understand and reproduce your stimulus selection without requiring you to republish protected images. Cite the relevant IAPS technical report or normative source, identify the specific image numbers used where permitted, and describe how pictures were grouped. For each condition, report the number of stimuli and descriptive statistics for the affective dimensions that matter to the hypothesis, such as mean and standard deviation of valence and arousal. Explain matching rules, exclusions, and whether you used sex-specific, population-specific, or local norms. State presentation duration, screen size or visual angle when relevant, interstimulus interval, randomization or counterbalancing, repetition, and the participant-rating procedure. If you modified image size, cropping, brightness, or color, report that because physical changes can affect perception. Also document ethics approval and the handling of potentially distressing stimuli where appropriate. In the results, distinguish normative stimulus values from participants’ observed ratings. Clear reporting helps reviewers assess whether an emotional effect reflects the intended manipulation rather than uncontrolled stimulus differences.