> ## Documentation Index
> Fetch the complete documentation index at: https://meta.niceshare.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Forbidden Fruit Effect

> Forbidden Fruit Effect is a higher wish to watch a violent program after a warning than after a content note. In 2006 adjusted means were 6.21 and 5.86.

<Info>
  **Category**: Effects<br />
  **Type**: A higher wish-to-watch rating for a violent program when the description adds advice, compared with a description that only names the content<br />
  **Origin**: Brad J. Bushman, *Journal of Applied Social Psychology* 36(9), 2006, pages 2073–2078. Peter Christenson, *Journal of Communication* 42(1), 1992, pages 106–113, named the rival pair. The Romeo and Juliet Effect is Driscoll, Davis, and Lipetz, 1972<br />
  **Also known as**: Romeo and Juliet Effect, the 1972 name for a parental-interference correlation with reported love. This page keeps that name on that correlation
</Info>

<Note>
  **Quick Answer** — The Forbidden Fruit Effect on this page is a higher rating of the wish to watch a violent television program when the description carries a warning than when it only says the program contains violence. Brad J. Bushman measured it in 2006 with 900 people. Adjusted means were 6.21 for the warning, 5.86 for the content note, and 5.68 with no label, on a scale from 1 to 10. The Romeo and Juliet Effect is a separate 1972 claim about parents and love. A 2014 repeat did not find a rise in love.
</Note>

## What is the Forbidden Fruit Effect?

The Forbidden Fruit Effect, as counted here, is a higher rating of "I would like to watch this TV program" when a violent description adds "Viewer discretion is advised" than when the description only says the program contains violence.

> Forbidden fruit theory posits that warning labels will increase interest in violent programs.

That sentence is Bushman and Stack's statement of the theory in 1996. The same abstract names the rival: tainted fruit theory says a warning will lower interest. The everyday picture is two notes on a jar. One note names what is inside. The other names what is inside and tells you to think twice. The rating that moved, in the television studies, is the second note. [Psychological Reactance](/effects/reactance) is the motive those papers cite, a push back when a freedom feels threatened. The rating on this page is the wish mark, not that motive's name.

### The Forbidden Fruit Effect in 3 Depths

* **Beginner**: In 2006 a warning sat above a content note. The adjusted means were 6.21 and 5.86 on a scale from 1, strongly disagree, to 10, strongly agree. The gap is under half a point.
* **Practitioner**: Write both sentences, and the no-label row, before you quote 6.21. The no-label mean is 5.68. The content note against no label was not a significant gap.
* **Advanced**: The Romeo and Juliet Effect is the 1972 name for a change in parental interference beside a change in love. In the 2014 repeat, that love correlation was −.05 and carried no significance star. Keep 6.21 on the television row.

## Origin

Peter Christenson, "The Effects of Parental Advisory Labels on Adolescent Music Preferences," *Journal of Communication* 42(1), March 1992, pages 106–113, is the paper Bushman and Stack cite for the two names. The journal's author note places him at Lewis and Clark College in Portland. The PDF was not re-read. The PsycInfo abstract says 145 people, ages 11 to 15, were assigned to a label condition or a control. Two unfamiliar albums, one hard rock and one pop, were the recordings, and the control group heard the same music without labels. The abstract says main effects for labeling and for musical type supported the tainted-fruit hypothesis. Pop was preferred to hard rock. Labeled music was liked less. No mean from that article is on this page.

Brad J. Bushman and Angela D. Stack, "Forbidden Fruit Versus Tainted Fruit: Effects of Warning Labels on Attraction to Television Violence," *Journal of Experimental Psychology: Applied* 2(3), 1996, pages 207–226, moved the question to television. The byline is Iowa State University. The PDF read here does not print a DOI. The publisher record is 10.1037/1076-898X.2.3.207. Experiment 3 used 180 undergraduate psychology students, 90 men and 90 women, 36 in each of five conditions, for extra course credit. They read adult film descriptions. A warning label or an information label sat under the violent ones, in uppercase and inside asterisks. The wish item used a 10-point scale from strongly disagree to strongly agree. Means adjusted for hours of violent television and for how interesting and exciting the films seemed were 6.1 for the warning conditions and 5.6 for the information conditions, F(1, 167) = 5.61, p \< .05, d = 0.37. The label effect overall was F(4, 167) = 2.49, p \< .05. The information-versus-no-label contrast was F(1, 167) = 0.04. The no-label mean is not printed. The other contrasts did not approach significance.

The counted means across ages are Bushman's 2006 paper, below. Jack Brehm's *A Theory of Psychological Reactance* (Academic Press, 1966) is the book Bushman cites for the motive. That book was not re-read. No toy count from it is on this page.

The Romeo and Juliet Effect is a different publication. Richard Driscoll, Keith E. Davis, and Milton E. Lipetz, "Parental Interference and Romantic Love: The Romeo and Juliet Effect," *Journal of Personality and Social Psychology* 24(1), October 1972, pages 1–10. The PDF was not re-read. The abstract says questionnaires went to 140 couples, and that parental interference intensifying romantic love was supported by cross-sectional correlations and by longitudinal change-score correlations. The publisher record is 10.1037/h0033373. No correlation from 1972 is on this page. H. Colleen Sinclair, Kristina B. Hood, and Brittany L. Wright, *Social Psychology* 45(3), 2014, pages 170–178, describe the original result as an increase in parental interference linked to an increase in love over the same 6-month period. Their own window was 3 to 4 months. They say they cut the original window in half.

## Key Points

Four readings keep each number on the row that produced it. A higher wish rating is a mark on a scale. A parental correlation is a different study.

<Steps>
  <Step title="Quote the warning row against the content row">
    In 2006 the warning was "This program contains violence. Viewer discretion is advised." The content note was "This program contains violence." Adjusted for hours of violent television, the means were 6.21 (SE 0.105) and 5.86 (SE 0.105). The contrast is t(869) = 2.35, p \< .05, d = 0.16. Three hundred people sat in each label condition. A slide that prints 6.21 as "people watched it" has left the sentence and the scale.
  </Step>

  <Step title="Keep the no-label row beside those two">
    The no-label mean was 5.68 (SE 0.104). Warning against no label was t(869) = 3.59, p \< .05, d = 0.24. Content note against no label was t(869) = 1.22, p > .3, d = 0.08. Tukey's test gave the warning its own subscript and gave the other two a shared subscript. The label effect was F(2, 869) = 6.65, p \< .005. The word "violence" is in both labeled sentences. The added advice is the contrast that cleared .05.
  </Step>

  <Step title="Set the 1996 choice test next to the rating">
    In Experiment 3 of Bushman and Stack, the wish rating was higher for warnings than for information labels, 6.1 against 5.6. The film people then started to play was not shifted by label, chi-square (4, N = 180) = 3.20, and none of the choice contrasts was significant. People who wanted the violent films more were more likely to pick one, rpb(178) = .47, p \< .05. The label still did not move the pick. [Streisand Effect](/effects/streisand-effect) is a later surge in attention after an attempt to hide a record. It is not this wish mark, and it is not this chi-square.
  </Step>

  <Step title="Keep Romeo and Juliet on the change-score row">
    Sinclair, Hood, and Wright followed 396 people and repeated the 1972 difference-score test. The correlation of an increase in parental interference with an increase in love was −.05 overall, −.13 among people who were dating, and .07 among people who were married. None of those three carries a significance star. An increase in commitment correlated −.19 with an increase in parental interference among daters, and that one is marked p \< .05. [Suspension Bridge Effect](/effects/suspension-bridge-effect) is arousal on a bridge, read as attraction. It is not this correlation.
  </Step>
</Steps>

## Applications

Write the sentence, the comparison sentence, and the scale before a wish rating leaves the table it came from.

<CardGroup cols={2}>
  <Card title="Write both sentences before 6.21" icon="list">
    The 6.21 is the warning row in Bushman 2006, adjusted for habitual violent viewing. The 5.86 is the row that already says the program contains violence. Quote them as a pair. A memo that keeps only 6.21 has dropped the comparison the study was built to make.
  </Card>

  <Card title="Ask whether anyone pressed play" icon="circle-play">
    The 2006 study collected a rating of a fictitious description. It did not record a finished viewing. The 1996 study did ask people to start a tape, and the label did not shift that choice. A plan that treats 6.21 as a play count has changed the outcome.
  </Card>

  <Card title="Leave the music sticker on its own page" icon="music">
    Christenson's 1992 abstract says labeled music was liked less. That is the tainted-fruit reading, in a group of 145 adolescents, with albums they could hear. [Scarcity Mindset](/thinking/scarcity-mindset) is a standing sense that there is not enough. A sticker on a record is neither that sense nor the 2006 television mean.
  </Card>

  <Card title="Keep a family argument on the 2014 row" icon="users">
    The 1972 abstract reports support for a rise in love alongside parental interference. The 2014 repeat, using the original measures, printed −.05 for the love change. An increase in friend approval did go with an increase in love, .38, marked p \< .001, on the newer scales. A household rule copied from the play has not repeated the television experiment.
  </Card>
</CardGroup>

## Case Study

In 2006 Bushman asked people ages 9 to 77 how much they wanted to watch fictitious, untitled made-for-television films that the cover story said had never been shown. The sample was 900: 180 in each of five age bands, half men and half women, with 300 in each label condition. Children came from public schools in central Iowa, with a parent's consent. The 18-to-20 group were introductory psychology students at Iowa State University. Adults who were not students, ages 21 to 77, mean 39 (SD 13), were approached in public places such as malls. Violent descriptions carried a warning, a content note, or no label. The warning added the sentence "Viewer discretion is advised."

The question was whether that extra sentence would raise the wish to watch. It did. Adjusted for hours already spent on violent television, the means were 6.21, 5.86, and 5.68. The warning stood above both other rows. The content note did not stand above the blank row. d = 0.16 is the warning-versus-content contrast. The paper says ratings of the nonviolent programs, which carried no label, were not influenced by which label the violent programs carried.

The lesson is narrow. Advice added to a content note raised a wish rating on fictitious descriptions, by less than half a point on a 10-point scale. The same paper's discussion says the warning drew viewers of all ages. The results section prints the label effect and a separate age effect. It does not print a label-by-age term, and Table 2's age means are averaged across the three labels. In 1996 the wish rating moved and the choice of which tape to start did not. A higher mark on "I would like to watch this" is the finding. A larger audience is a further claim this design did not score.

## Boundaries and Failure Modes

The Forbidden Fruit Effect has clear boundaries:

**Boundary 1: A music sticker went the other way.** Christenson's abstract, the source Bushman and Stack cite for the two theory names, says labeled albums were liked less. The television warning and the record sticker are different objects, different ages, and different outcomes. One abstract does not lend its direction to the other.

**Boundary 2: Parental interference did not raise love in the repeat.** Sinclair, Hood, and Wright analyzed 396 usable responses from an online sample, ages 18 to 70, mean 31.58 (SD 9.78), of whom 48.5 percent were married. Their introduction says the 1972 couples showed a rise in love over 6 months. Their Table 2 does not show that rise. Increases in parental interference did go with lower trust, −.18, and higher criticism, .22, both marked p \< .001 in the full sample. On the newer scales, an increase in parental approval went with an increase in love at .24, marked p \< .001, while the married row of that pair is .02 and has no star. A meta-analysis of 22 studies, including theirs, printed an overall Hedges' g of 0.49 for network approval and love (11 effects, 95 percent CI 0.26 to 0.72) and 0.62 for commitment (16 effects, CI 0.50 to 0.74). The love row also prints I² = 53.92. Approval, in that pool, sits with higher love. The studies are not one size.

**Common misuse**: Reading 6.21 as proof that banning a song, a romance, or a product will raise demand. The 6.21 is one cell of a rating of fictitious television descriptions. The 1992 abstract and the 2014 change scores are the nearby rows that do not match that story. [Backfire Effect](/effects/backfire-effect) is a different argument about a correction that strengthens a belief. It does not inherit these means.

## Common Misconceptions

The name travels farther than the tables. Each claim below has a row that already answers it.

<AccordionGroup>
  <Accordion title="Misconception: Any ban makes people want the thing">
    **Reality**: In 2006 the content note, which already named the violence, did not differ from no label. In 1992 the abstract says a parental-advisory sticker went with lower liking for the music. A ban, a sticker, and a sentence of advice are three different restrictions. Only the last one is the contrast that raised the 2006 wish rating.
  </Accordion>

  <Accordion title="Misconception: The Romeo and Juliet Effect means opposition deepened love">
    **Reality**: That is the claim in the 1972 abstract. Sinclair, Hood, and Wright repeated the difference-score test and did not find a positive link between rising parental interference and rising love. The printed love correlations are −.05, −.13, and .07, with no significance stars. Among daters, rising interference went with a drop in commitment, −.19, marked p \< .05. The paper uses Felmlee's 2001 phrase, the social network effect, for the pattern it did find: more interference, or less approval, with poorer relationship quality. Felmlee's article was not re-read. No breakup rate from it is on this page.
  </Accordion>

  <Accordion title="Misconception: The wish rating is a count of viewers">
    **Reality**: Both television studies asked for agreement with a wish sentence. The 2006 means sit between 5 and 7 on a scale whose 10 is strongly agree. The 1996 choice test was not significant. A program that more people say they would like, on a questionnaire about a film that does not exist, has not been shown to draw a larger audience.
  </Accordion>
</AccordionGroup>

## Related Concepts

The wish rating sits beside a motive, a surge in attention, a sense of shortage, a misread arousal, and a correction that strengthens a belief.

<CardGroup cols={3}>
  <Card title="Psychological Reactance" icon="brain" href="/effects/reactance">
    The motive Bushman cites when advice feels like a limit on what a person may watch. Brehm's 1966 book was not re-read. The 6.21 is a television rating, not a score on a reactance scale.
  </Card>

  <Card title="Streisand Effect" icon="eye" href="/effects/streisand-effect">
    Attention that grows after an attempt to suppress a record. The 1996 chi-square is a choice among tapes in a lab. It is not a later public audience.
  </Card>

  <Card title="Scarcity Mindset" icon="scale-balanced" href="/thinking/scarcity-mindset">
    A standing sense that there is not enough. The 2006 contrast is two sentences under a fictitious description, adjusted for hours of violent television.
  </Card>

  <Card title="Suspension Bridge Effect" icon="bridge" href="/effects/suspension-bridge-effect">
    Arousal on a bridge, later read as attraction to a person. The Romeo and Juliet row on this page is a correlation of interference with love, and the 2014 love change is −.05.
  </Card>

  <Card title="Backfire Effect" icon="fire" href="/effects/backfire-effect">
    A correction that is said to strengthen the belief it challenges. A warning label on a television description is a different sentence, and 6.21 is not that correction.
  </Card>
</CardGroup>

## One-Line Takeaway

<Tip>
  Before you treat a warning as a rise in demand, write the comparison sentence, the scale, and whether anyone pressed play.
</Tip>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.