Guidelines on Thoughtful Image Selection for Instructors

• Purpose: Why are you engaging in this work around image selection?
• Task: What do you hope comes out of this work around image selection?
• Criteria: How do you plan to go about your work around image selection?

Identify academic community members from diverse backgrounds who may be interested in joining your team and provide advice about the images that you are going to create and use. If possible, include members from your campus partners, so that individuals are not being asked to speak as a representative of their group to confirm an image.

Seek input from academic community members about the main messages that you want to convey and the types of images that would be most suitable. Be mindful of the pre-existing conversations happening on the topic in other departments that could inform or influence the approach. Create spaces to process and host conversations around image selection to ensure new and multiple perspectives are being reflected, including the perspectives of those who might be portrayed in the image.

Share draft images with your team for review and feedback. It is important to leave room for informal and formal conversations to honor the variety of ways that images can be created and interpreted.

The Coalition for Diversity & Inclusion in Scholarly Communications suggests that the images in an entire project should be examined to identify any problematic patterns (see Inclusive Images).  The authors point to problematic patterns identified by King & Domin (2007), such as images of people of color as typically being shown in a less visible or important sidebar instead of the main feature. In other words, carefully observe and consider how you center people and lived experiences.

Revisit this process as new images are selected or updated, and when staff and roles change. Continue to check in with campus thought partners as this allows opportunities for staying up to date on current topics that might inform image selection.

Relevance of the Image: Choose images that are directly relevant to the content. If the image does not support or enhance the message, consider omitting it.

Human vs. Non-Human Images: An international study of digital images (Cyr et al., 2009) conducted in the United States, Japan, and Germany demonstrated that images of human faces created more engagement with audiences, including increased time looking at an image. Despite increased engagement, images of people may not always be the best choice because of the inherent limitations in human representation.

Using images of people—including photographs, illustrations, and abstract representations—should be done conscientiously, recognizing that non-human images may be the best option in certain contexts. 

• Images of people can create an emotional connection with audiences. 
• Non-human images like landscapes, icons, or abstract designs are useful for general topics and concepts and for highlighting technology, infrastructure, or locations.

Images Created by Artificial Intelligence (AI): Use care when creating images with AI. A feature article published in Nature in 2024 presents evidence that images generated by AI are prone to gender and racial bias. The use of AI to create images is still contested as it does not acknowledge the sources or creators that the AI draws from to produce images.

Diverse Representation: Ensure that the images visibly reflect diversity in race, gender, age, and abilities. Aim for a balanced representation that mirrors the Northwestern community or the community that you are working alongside—a key feature of embracing the participatory and collaborative nature of image selection.

Stereotypes: Choose images that challenge clichés and stereotypes rather than reinforcing them. Be mindful of the roles and settings in which people are portrayed. Be mindful about the lived experiences that may be conveyed or omitted by an image and avoid images that reinforce stereotypes or reduce a culture to a single aspect or trait. This includes images that portray a culture in a superficial or inaccurate manner.

Killpack and Melón (2017) demonstrate that everyone is susceptible to exacerbating stereotype threat and confirming negative stereotypes about a particular group. Selecting images that challenge or avoid stereotypes can help to avoid this exacerbation and confirmation (Steele, 1997).

Cultural Relevance: Be mindful of the cultural implications of images. Ensure that cultural elements are depicted respectfully and accurately (see Cultural Appropriation).

Local vs. Global: Decide when to use local imagery that resonates with the immediate community versus when to use images that appeal to a global audience.

Cultural Appropriation: Cultural appropriation occurs when elements of a marginalized or minoritized culture are taken and used by members of a dominant culture, often without understanding, respect, or permission. This can include the use of traditional attire, symbols, music, language, and rituals in ways that strip them of their original meaning or context. To avoid cultural appropriation in selecting images, it is important to:

• Recognize the significance of the cultural element being depicted. Understand that cultural elements carry deep significance and history. Using them in ways that do not honor their origin can be harmful and disrespectful to the community from which they originate. 
• Ensure authentic representation. When using images that depict diverse elements of culture, choose those that authentically represent the culture. This can often be achieved by using images created by members of that cultural community or by involving them in the selection process. 
• Seek input and feedback. Whenever possible, obtain consent and input from members of the cultural group being represented. This helps ensure that the representation is both accurate and respectful.

Resolution: Use high-resolution images to avoid pixelated or blurry images, unless it is part of the artistic intent, which should be clearly communicated in alt text (see ACCESSIBILITY: Alt Text).

Consistency: Opt for a consistent visual style. This could relate to the color scheme, level of formality, or the type of images selected.

Alt Text: Provide alternative text (alt text) for all images to ensure that the content is accessible to individuals using screen readers. The Northwestern office for accessibility, AccessibleNU, defines alt text as text that “describes the content of images, graphs and charts”. Refer to AccessibleNU’s guidelines for how to label images for alt text. Free online resources, such as Arizona State University’s Image Accessibility Creator, use the power of AI to create alt text for simple and complex images.

• To consider: Different fields have different guidelines for alt text. For example, journals may require very specific text and numbers of characters for labeling figures and tables.

Contrast: Good color contrast supports visibility and accessibility. Free online resources, such as Colour Contrast Checker, allows users to check color contrast, typeface, and font size for accessibility.

Color Palettes: Choose accessible color palettes. There are many free online resources, such as Venngage’s Colorblind-Friendly Palettes, that provide colors and color palettes that are accessible for people who are color blind. If colors and color palettes are limited by branding guidelines, use the color plus approach.

• Color Plus Approach: David Nichols, creator of the online colorblindness simulation tool, suggests avoiding conveying information purely through color, because using only mode of communication limits accessibility for those who face challenges with that mode. He suggests using the color plus approach, adding other means such as text, symbols, or patterns to present information in multiple forms.

Canvas: Ensure accessibility in course site content for students and users. Canvas Style by Pennsylvania State University is a resource that addresses several areas of formatting content on the learning management system, with accessibility in mind, like images, tables, and videos in Canvas.

Permissions and Rights: Only use images that are available for public use or that you have received permission to use. This includes understanding rights for stock photos, copyrighted material, and personal photographs. (See Image Sourcing.)

Respect Privacy: Obtain consent from individuals whose faces are recognizable in the images. Northwestern has policies and consent forms for sourcing photographs.

Transparency: As a responsible community member and user, be explicit about the ownership of imagery and use of imagery in ways that cite and give credit to a creator for an original image. Reach out and communicate for consent, including an option to opt out.

Stock Images: Northwestern faculty and staff have access to 2 libraries of stock images: Adobe Stock and Northwestern's Marketing Digital Asset Management System (MDAM). MDAM provides a wide selection of images that are particular to Northwestern.

• Adobe Stock can be accessed with a Creative Cloud account. Follow NUIT prompts provided here to set up an account, if you do not have one.
• MDAM can be accessed with your NetID. Follow prompts provided here to access the system. 
• Both Adobe Stock and MDAM have filters that allow you to find images that align with your needs and values. For instance, in Adobe Stock, you can filter image results to show those created by local artists or to exclude images created by generative AI. In MDAM, you can double check the usage rights for each image.