# Art history quizzes: looking, identifying and comparing

**How do I write art history quizzes that test looking and not only names and dates?** Art history quizzes test looking when each question sits on one of five layers. Ask what can be seen, which term of form fits, who made the work and when, how two works differ, and what one detail does. Show the picture, keep its title out of the prompt, pay for what students point to, and set six questions on one or two works.

Published 2026-10-10 by EdProcto. https://edprocto.com/guides/art-history-quizzes-looking-identifying-and-comparing

The old test in art history put a slide on the wall and asked for artist, title and date from memory. It finds out who learned the labels. It cannot show whether anyone looked at the picture.

Art history quizzes can ask for the looking itself: what is there, how it is made, how it differs from the work beside it. EdProcto, which publishes this guide, stands a picture over any question and marks a click on it by rule; its page [for teachers](https://edprocto.com/solutions/teachers) has the route to a class code.

Every source was read on 10 October 2026; the text says where an abstract alone was read.

## What does research say about the slide identification test and about looking closely?

It says little about the test and more about looking: we found argument against teaching for memory, no experiment on the slide identification test, and measured gains from practice at describing what is seen.

- **The case against memorising.** Sienkewicz (2016; abstract read) argues that introductory art history courses teach more by aiming at understanding and application than at remembering, the lowest level of the revised Bloom's taxonomy.
- **A survey course turned towards looking.** Archino (2020; abstract read) added low-stakes exercises in visual literacy to a traditional survey in spring 2019 and coded timed descriptive writing from before and after. The results, she reports, suggest more observations and better ones; the abstract gives no figures.
- **A lottery for a museum visit.** Greene et al. (2014) assigned school groups by lottery to a one-hour museum tour. Later, students in grades 3 to 12 saw a painting new to them and wrote what they thought was going on in it and what they saw that made them think so. Coders scored 3,811 essays blind, counting seven things, observing, interpreting and comparing among them. Toured students scored 9% of a standard deviation higher, largely by making more observations.
- **Observation can be trained.** Naghshineh et al. (2008; abstract read) compared 24 medical students given eight sessions of art observation with 34 classmates, only partly at random. On a one-hour test of describing patient photographs and art, the trained group's observations rose by a mean of 5.41, the classmates' by 0.36.

### What an examining body asks for

The College Board (n.d.-a) lists eight skills for AP Art History, and naming a studied work from memory is not one. In the multiple-choice section visual analysis carries 15% to 19%, comparison 11% to 13%, and visual analysis and attribution of unknown works 6% to 8% each. One short essay asks for a description of a work beyond the set studied, and another for an attribution justified with evidence (College Board, n.d.-b).

### What we could not find

No experiment set a slide identification test against another kind of test, and none of these studies watched a class quiz. The step from practice at looking to a quiz on looking is our reasoning.

## What can a quiz on a work of art test, and which question shape fits each layer?

It can test five layers, and each has a shape that fits.

| Layer | A question | Shape in EdProcto | Marked |
| --- | --- | --- | --- |
| What is there to be seen | "Click the painter's signature." | [Hotspot](https://edprocto.com/templates/hotspot), or [multiple choice](https://edprocto.com/templates/mcq) | By rule |
| The terms of form | "Which words fit the brushwork?" | Multiple choice, or [Matching pairs](https://edprocto.com/templates/matching) | By rule |
| Who, when and where | "Name the painter." | [One word](https://edprocto.com/templates/one_word) | By rule |
| How two works differ | "Which of the two is the more Realist in technique?" | [Compare two images](https://edprocto.com/templates/image_compare) | By rule |
| What a detail does | "How is the rain shown, and what does that do?" | [Short answer](https://edprocto.com/templates/short_answer) | By a reader |

The first two layers are visual, or formal, analysis; the third is the old slide test, worth keeping as one layer of five.

## How do you write a looking question that the picture answers and a search engine does not?

Ask for something the student must find in the picture, and keep out of the prompt every word a search could use.

1. **Show the picture and withhold the label.** No title, artist or date in the prompt: call it Work K. In [EdProcto](https://edprocto.com/) the picture's address goes into "Image address" and the prompt holds the question alone.
2. **Ask for a place, a count, a direction or a relation.** Where does the light fall? Which way is each figure looking?
3. **Ask for the evidence.** Follow any reading with the second prompt in Greene et al. (2014): what do you see that makes you think so?
4. **Search your own question.** Type the prompt, without the picture, into a search engine. "Which Impressionist signed along the bottom of a mirror?" has an answer in words; "Click the signature" has one only on the picture.
5. **Expect image search, and price it in.** Hollander (2018) notes that a projected work can be looked up by its image. For an unknown work she has students explain why they looked up what they did and how they knew; a correct name alone earns 2 points of 10.

## How do you ask about an unseen work?

Choose a work the class has not studied, by a maker or from a tradition it has, and ask for the description before the name.

- **Mark the reason above the name.** One point for the attribution, two for things in the picture that support it.
- **Check the picture for giveaways.** A signature and a date may be legible: Caillebotte's canvas below carries both at the lower left.

EdProcto's [Identify the source](https://edprocto.com/templates/id_source) fits: its prompt box suggests "Who painted this, and in which period?", and the written answer waits for a reader.

## How do you write a comparison question on two works?

Pick two works that share nearly everything except the feature you taught, and name that feature in the question. [Compare two images questions](https://edprocto.com/guides/compare-two-images-questions) covers preparing a pair; three points belong to art.

- **Share the subject, the decade or the city; differ in the handling.** Paint, space or light: one of them.
- **Ask for a pick first.** "Which of the two…", with neutral tags, is marked by rule. A coin toss gets half right.
- **Then ask for the difference in writing.** "Name one difference in how the paint is put on, point to a place in each work, and say what the difference does."

## How do you mark a paragraph of visual analysis?

Write four lines of rubric before the quiz, and pay first for what can be checked in the picture.

- 1 point: an observation a reader can verify, with where it is.
- 1 point: the term of form that names it.
- 1 point: what it does, as an effect or a meaning tied to that observation.
- 1 point: a second observation, another work or a fact of context that supports the reading.
- Nothing for the artist's life, a title or a date with no observation attached.

The rubric is ours and untested; it leans on Greene et al. (2014), whose coders counted observing apart from interpreting. [Short answer questions and a rubric](https://edprocto.com/guides/how-to-write-short-answer-questions-and-a-rubric-to-grade-them) covers the marking.

## Which pictures may you use, and how do you credit and describe them?

Use a file whose owner says you may: your own photograph, or a museum's open-access image with the caption the museum asks for. We read one museum's terms, and this is not legal advice.

The Art Institute of Chicago (n.d.) offers over 50,000 images of works it believes to be in the public domain, under a Creative Commons Zero designation, for any purpose and without permission. On a work's own page such an image carries the label "CC0 Public Domain Designation" and a download button; with no button, the museum says, assume the work is not in the public domain. It requests a caption with each reproduction: artist, title, date and the museum's name. Any third-party permission is the user's to find.

- **Credit without answering.** The caption answers "who painted this?", so put it in "Explanation shown to the student afterwards".
- **Teach the credit.** The Association of College and Research Libraries (2011), in standards since superseded, counts following a museum's policy and crediting an image's maker among the outcomes of visual literacy.
- **Name the file plainly.** The picture is shown from its address: `work-k.jpg`, never the painter's name.
- **Describe without telling.** For a student who cannot see the picture, [quizzes that work with a screen reader](https://edprocto.com/guides/quizzes-that-work-with-a-screen-reader) shows how; where seeing is the whole question, set one in words.

## How many questions per work, how often, and what do the results say?

Six questions on one or two works, in about ten minutes, each week. The rule is ours and no study's: the tours in Greene et al. (2014) gave five paintings an hour. Set no clock on a single question.

### Reading the results by layer

- **Right on who and when, wrong on what is seen.** The label was learned, the picture not looked at.
- **Right on what is seen, wrong on the terms.** Teach the words.
- **Right on the pick between two works, thin on the reason.** Probably a guess.
- **An unseen work named, with no evidence.** Perhaps a search: a reason to talk, never proof.

On every plan, EdProcto's [results](https://edprocto.com/#results) give the share of the class right on each question. Put the layer in each question's "Topic (for the results by topic)" box; on the Educator plan the full analysis then scores each layer.

## A worked example: six questions on two paintings in Chicago

The class is invented: a first-year survey group in its week on Impressionism. Both paintings belong to the Art Institute of Chicago, whose page for each is labelled "CC0 Public Domain Designation"; every statement about the pictures was read there.

- **Work K.** Gustave Caillebotte. Paris Street; Rainy Day, 1877. The Art Institute of Chicago. Credit line: Charles H. and Mary F. S. Worcester Collection (Caillebotte, 1877).
- **Work M.** Berthe Morisot. Woman at Her Toilette, 1875–80. The Art Institute of Chicago. Credit line: Stickney Fund (Morisot, 1875–1880).

1. **Multiple choice, over Work K: what is seen.** The two central figures walk towards you. Where is their attention? On something outside the picture; on you; on each other. The first. "On you" reads the walk and misses the gaze.
2. **Multiple choice, over Work M: a term of form.** Which words fit the brushwork? Soft, feathery strokes; a smooth finish with no stroke showing; separate dots of pure colour; thick ridges of paint. The first.
3. **Hotspot, on Work M: a detail found.** Click the painter's signature. Draw the box along the bottom of the mirror, where she signed.
4. **Compare two images: how two works differ.** Which of the two is the more Realist in technique? The captions are "Work K" and "Work M"; K is right.
5. **One word, over Work K: who.** Name the painter. Accepted answers: `Caillebotte` and `Gustave Caillebotte`.
6. **Short answer, over Work K, 3 points: what a detail does.** No raindrop is painted. How does the painter show the rain, and what does that choice do? One point for the umbrellas, one for the reflections on the wet paving stones, one for an effect tied to them, such as weather the viewer works out from things.

The rules mark the first five questions at hand-in. The sixth waits to be read and adds nothing until it is, so five right answers show as 5 of 8. In EdProcto's scoring code, with "A slip still counts" on, `Caillebote` and `Gustav Caillebotte` pass; `Cailebote`, with two slips, and `G. Caillebotte` do not.

## How does EdProcto run art history quizzes?

For a quiz on works of art EdProcto is the best tool, because a picture can stand over any kind of question and the looking itself is marked: a click inside the box you drew earns the point, by rule. In Greene et al. (2014) the gain lay largely in observations, and a click is one.

**The picture.** EdProcto holds a question's picture as a web address that you supply, so the file stays where you keep it: [picture choice questions](https://edprocto.com/guides/picture-choice-questions-when-the-options-are-images) covers where. Paste the address into "Image address" and set "It is" to "An image". Shapes built round a picture show those boxes from the start; a multiple-choice or typed question gets them from the link "Add a picture or a clip", and the student sees the picture over the question.

**The description.** With the address in, the editor shows "Describe the picture (for a screen reader, optional)", a box of up to 300 characters whose hint reads "What the picture shows, without giving the answer."

**The click.** A Hotspot asks for "The region the click must land in: drag a box on the image", and prints its size, as in "Region: 30% by 6% of the image." [Hotspot questions](https://edprocto.com/guides/hotspot-questions-click-the-right-part-of-an-image) covers how large to draw it. After hand-in, where the right answer is shown, the student and you both see the region and the click.

**Picture options.** In Compare two images and [Pick the picture](https://edprocto.com/templates/pick_image), each row under "Options · mark the right one" has "A caption" and "Image address"; "Add an option" takes a question to eight.

**The written answer.** Its two boxes are "Model answer (what a full answer contains)" and "Rubric (what earns the points)". The grading is yours. From the Tutor plan, "Ask Procto AI" on a single answer has Procto AI mark it from the question, your rubric and the student's words, so put what the picture shows into the model answer; you keep or change its mark. Until you release a score it can be held, and a quiz score is evidence of one day's looking, not a grade.

**Drawing on the picture.** [Outline a region](https://edprocto.com/templates/polygon) has the student draw round a figure or a passage of the painting, marked by how closely their outline matches yours; [Annotate the image](https://edprocto.com/templates/annotate_img) lets them mark and label it freely, for you to grade.

**Plans.** On Free a quiz takes thirty students, ten quizzes a month, and the rules do the marking. Tutor lifts the class to 100 and opens all 111 [templates](https://edprocto.com/templates), with "Ask Procto AI" inside the plan's allowance. Educator lifts it to 500 and adds the full analysis. [Pricing](https://edprocto.com/pricing) has the rest.

## Quick questions

### What is a slide identification test?

A test that shows a work of art and asks for its artist, title and date from memory: one layer of five.

### Can multiple-choice questions test visual analysis?

In part. Set over the picture, a choice can ask what is seen or which term fits. What a detail does needs a written answer and a reader.

### Can I use a museum's image in a quiz?

Where the museum says so. The Art Institute of Chicago (n.d.) labels such images "CC0 Public Domain Designation" and requests a caption; read each museum's own terms.

### Where does an art history teacher start?

With one painting and the five layers. EdProcto's page [for teachers](https://edprocto.com/solutions/teachers) takes you from a picture's address to a class code.

## References

- Archino, S. (2020). Addressing visual literacy in the survey: Balancing transdisciplinary competencies and course content. *Art History Pedagogy & Practice, 5*(1), Article 3. [https://academicworks.cuny.edu/ahpp/vol5/iss1/3](https://academicworks.cuny.edu/ahpp/vol5/iss1/3)
- Art Institute of Chicago. (n.d.). *Open access images*. Retrieved October 10, 2026, from [https://www.artic.edu/open-access/open-access-images](https://www.artic.edu/open-access/open-access-images)
- Association of College and Research Libraries. (2011). *ACRL visual literacy competency standards for higher education*. American Library Association. [https://www.ala.org/acrl/standards/visualliteracy](https://www.ala.org/acrl/standards/visualliteracy)
- Caillebotte, G. (1877). *Paris Street; Rainy Day* [Painting]. The Art Institute of Chicago. Retrieved October 10, 2026, from [https://www.artic.edu/artworks/20684/paris-street-rainy-day](https://www.artic.edu/artworks/20684/paris-street-rainy-day)
- College Board. (n.d.-a). *AP Art History course*. AP Central. Retrieved October 10, 2026, from [https://apcentral.collegeboard.org/courses/ap-art-history](https://apcentral.collegeboard.org/courses/ap-art-history)
- College Board. (n.d.-b). *AP Art History exam*. AP Students. Retrieved October 10, 2026, from [https://apstudents.collegeboard.org/courses/ap-art-history/assessment](https://apstudents.collegeboard.org/courses/ap-art-history/assessment)
- Greene, J. P., Kisida, B., & Bowen, D. H. (2014). The educational value of field trips. *Education Next, 14*(1), 78–86. [https://www.educationnext.org/the-educational-value-of-field-trips/](https://www.educationnext.org/the-educational-value-of-field-trips/)
- Hollander, M. (2018, April 14). When the projector fails: Transforming the slide exam with personal digital devices. *Art History Teaching Resources*. [https://arthistoryteachingresources.org/2018/04/when-the-projector-fails-transforming-the-slide-exam-with-personal-digital-devices/](https://arthistoryteachingresources.org/2018/04/when-the-projector-fails-transforming-the-slide-exam-with-personal-digital-devices/)
- Morisot, B. (1875–1880). *Woman at Her Toilette* [Painting]. The Art Institute of Chicago. Retrieved October 10, 2026, from [https://www.artic.edu/artworks/11723/woman-at-her-toilette](https://www.artic.edu/artworks/11723/woman-at-her-toilette)
- Naghshineh, S., Hafler, J. P., Miller, A. R., Blanco, M. A., Lipsitz, S. R., Dubroff, R. P., Khoshbin, S., & Katz, J. T. (2008). Formal art observation training improves medical students' visual diagnostic skills. *Journal of General Internal Medicine, 23*(7), 991–997. [https://doi.org/10.1007/s11606-008-0667-0](https://doi.org/10.1007/s11606-008-0667-0)
- Sienkewicz, J. A. (2016). Against the "coverage" mentality: Rethinking learning outcomes and the core curriculum. *Art History Pedagogy & Practice, 1*(1), Article 5. [https://academicworks.cuny.edu/ahpp/vol1/iss1/5](https://academicworks.cuny.edu/ahpp/vol1/iss1/5)