This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Youths eat fewer fruits and vegetables than recommended. Effective methods are needed to increase and maintain their fruit and vegetable consumption. Goal setting has been an effective behavior change procedure among adults, but has had limited effectiveness among youths. Implementation intentions are specific plans to facilitate goal attainment. Redefining goal setting to include implementation intentions may be an effective way to increase effectiveness. Video games offer a controlled venue for conducting behavioral research and testing hypotheses to identify mechanisms of effect.
This report describes the protocol that guided the design and evaluation of Squire’s Quest! II, a video game aimed to increase child fruit and vegetable consumption.
Squire’s Quest! II is a 10-episode videogame promoting fruit and vegetable consumption to 4th and 5th grade children (approximately 9-11 year old youths). A four group randomized design (n=400 parent/child dyads) was used to systematically test the effect of two types of implementation intentions (action, coping) on fruit and vegetable goal attainment and consumption of 4th and 5th graders. Data collection occurred at baseline, immediately post game-play, and 3 months later. Child was the unit of assignment. Three dietary recalls were collected at each data collection period by trained interviewers using the Nutrient Data System for Research (NDSR 2009). Psychosocial and process data were also collected.
To our knowledge, this is the first research to explore the effect of implementation intentions on child fruit and vegetable goal attainment and consumption.
This intervention will contribute valuable information regarding whether implementation intentions are effective with elementary age children.
ClinicalTrials.gov NCT01004094
Consuming adequate amounts of fruit and vegetables (FV) is part of a healthy lifestyle [
Video games are popular among youths [
Serious video games, ie, video games designed to entertain as well as achieve change of some type [
Implementation intentions are specific plans that identify how to achieve a goal [
Long-term behavior change is the ultimate goal for behavioral interventions [
Squire’s Quest! II: Saving the Kingdom of Fivealot (SQ!2) is a 10-episode online video game designed to increase FV consumption among 4th and 5th grade children (roughly 9-11 year old youths). It is an updated and enhanced version of an earlier video game, Squire’s Quest!, evaluated in 1999 and 2000 [
This evaluation used a four-group, randomized design, with three data collection periods (baseline, post 1, post 2). Following baseline assessment, children were randomized to one of four groups: goal setting only (simple), goal setting + action intentions (action), goal setting + coping intentions (coping), or goal setting + action and coping intentions (both). Youths had up to three months to play all 10 episodes of the video game, where the appropriate goal-setting/implementation intentions were embedded in four versions of the game. Post 1 data collection occurred immediately upon completion of the 10 episodes or approximately 3 months after beginning game-play, whichever occurred first. Post 2 data collection occurred approximately 3 months after post 1 (
Research design.
Eligibility criteria for participants included: (1) a child in the 4th or 5th grade at time of enrollment, (2) fluent in English, (3) had access to a computer with high speed Internet, and (4) had a parent (or legal guardian) fluent in English or Spanish who was willing to participate in the study. Recruitment methods included standard procedures (ie, flyers and attendance at community events) and the volunteer database at the Children’s Nutrition Research Center. Prior to participation, eligible parents and children provided written informed consent and child assent. The child was the unit of randomization.
Sample size requirements were based on a power analysis (the number of participants needed to find an actual group difference) with FV as the primary outcome variable. A repeated measures analyses of variance with a 4-group design and 3-measurment periods was assumed. Allowing for a 30% attrition rate, a sample of 400 parent/child dyads (100 per group) provided adequate power (>80%) to detect a small effect size (ES=0.17) using a standard deviation of 1.5 from the original Squire’s Quest! and a two-sided alternative with type I error rate of 0.05. This effect size (ES=0.17) translates to a detectable 0.51 or greater serving (ie, a half of a serving per day) group difference.
Four hundred multi-ethnic parent/child dyads were enrolled in the study. Fifty three percent of children were girls, while 47% were boys; racial/ethnic distribution was 37% White, 27% Hispanic, 26% Black, and 10% Other. Most parents were female (96%), and parent racial/ethnic distribution was similar to that of the children (41% White; 26% Hispanic; 26% Black; 7% Other).
Parents and children participated in separate intervention and data collection activities electronically (Internet or telephone) from locations of their choice (home or community). There were no face-to-face sessions.
The intervention had both child and parent components. The child played a 10-episode online video game, while the parent received electronic newsletters and access to a parent website which was updated with new information10 times, corresponding to the 10-episode video game. The intervention was guided by a theoretical framework that incorporated social cognitive (behavioral and environmental factors) [
Conceptual model.
Multi-ethnic 4th and 5th grade children provided information for the development and initial testing of the video game. Children (n=15) provided feedback on behavior change components, the storyline, and the characters. Children (n=10) also participated in alpha and beta testing sessions. An experienced 4th grade teacher reviewed materials to assess developmental appropriateness and comprehension of key components. Children and their parents (n=20 dyads) also participated in a pilot study, which served as a final test for the intervention components, procedures, and data collection activities of the video game.
The story, written by a professional writer, was designed to appeal to 4th and 5th grade youths (ie, humor appropriate for children). It integrated the behavior change components into the action, which, in turn, advanced the storyline. The “backstory” (ie, the history) provided context for the storyline.
The medieval kingdom of Fivealot is a happy one; its prosperity is based on ancient knowledge of healthy nutrition and an abundance of good food, under the gentle guidance of King Brocwell and Queen Nutritia. Recently, Fivealot entered a truce with their underground-dwelling enemy, the Mog – or so they thought.
A nation of subterranean serpents and myopic moles, the Mog were jealous of Fivealot’s good fortune. Although Brocwell and Nutritia had offered to share their country’s bounty, the Mog who were raised on a diet of sweets and fried foods, stubbornly refused to change their ways. The Mog King Snake, Sssynster, instead insisted that Fivealot change
Discovering that their army of fattened moles and slovenly snakes were no match for the energetic knights of Fivealot, Sssynster withdrew, realizing he could never best Brocwell and Nutritia in open conflict. His best hope was treachery.
The King and Queen of Fivealot’s Head Chef, Supremo, was preparing to go to market one day. With the truce in effect, he suspected nothing – certainly not the tunnel dug into his kitchen by the Mog. Unbeknownst to Brocwell and Nutritia, Supremo simply disappeared, spirited away to the land of Mog.
Then came the second part of Sssynster’s plan: a double agent, Moledred, had been placed on the kitchen staff, waiting for the opportunity to advance to the Head of the Kitchen. With the majority of knights away spreading the bounty of Fivealot, Moldred’s mission was to prepare his specialty - fattening, unhealthy meals - for the King and Queen, and thus bring the kingdom to ruin. But what Sssynster and Moledred had not counted on was a Squire answering the King and Queen’s call for brave young men and women to train in the ways of Fivealot and become knights themselves. A Squire…like you.
To become a knight, the Squire had to acquire the coveted knowledge and skills of the Fivealot Knights. This is where the game began. In the quest for knighthood, the Squire had to overcome challenges. The challenges involved attaining “real world” FV consumption and recipe goals. As the Squire (ie, the participant) met their challenges, they earned badges and progressed in their journey towards knighthood. Because children typically consume well below the recommended level of daily FV [
The protagonists included six characters that were “human” in appearance (King Brocwell; Queen Nutritia; Merlin the Wizard; Knights Alex and Julie; Chef Supremo) and a robot (M.I.C.H.A.E.L.), who assisted with kitchen tasks. The antagonists were snakes (King Sssynster, a cobra; the snake army) and moles (Moledred, the imposter chef) (
Characters.
The genre was action adventure. The game contents and flow diagram (ie, the video game structure) are presented in
Game flow diagram.
Players were assigned a unique username and password with which to login to the game. After completing baseline data collection, players were randomly assigned to one of four versions of the game. Their username automatically routed them to their assigned version. The only differences between the groups were the implementation intentions described in the goal review and goal setting sections below.
Goal review appeared in episodes 2-10. It occurred at the beginning of each episode and was led by the Wizard. Players reported whether they met the two types of goals set during the previous episode: a FV consumption goal and a FV recipe preparation goal. In the video game, goals were referred to as challenges to be consistent with the “quest” towards knighthood. Players in one of the three groups that created an implementation intention reported whether they followed it. Players, regardless of group assignment, also reported whether they used the skills learned in the video game (ie, self monitoring, problem solving, and asking/negotiation) to help them meet their goals. For each unmet goal, the player participated in a problem solving sequence to identify the problem that kept them from meeting their goal. Players received feedback statements (tailored to their level of success in meeting their goals and/or whether they used skills to help them meet their goals). Players were then asked whether the effort they put into eating FV was worth it (an assessment of satisfaction). To reinforce goal commitment, they were then encouraged to type in a positive self statement (eg, “Setting and meeting my challenges shows I’ve got what it takes to be a winner!”) and read it out loud with conviction.
Badges were awarded in episodes 2 - 10 for meeting the goals set in the previous episode. Players could earn up to two badges each episode, one for meeting the FV goal and one for meeting the FV recipe goal. Total number of badges earned determined level of knighthood. There were five levels, ranging from the lowly Honorary Knight to the coveted Platinum Knight. Badges appeared on the player’s coat of arms (or shield) displayed in the castle foyer (
Shield with badges.
The cut scenes were animated video clips that presented the story. The story was told from the second person perspective. The cut scenes provided an opportunity for the characters to serve as role models through dialogue and action. To make the player feel “part of the action”, the characters “spoke” to the player by referring to him/her as “Squire”.
This component appeared in episodes 1 - 10. It presented basic knowledge [
Knowledge topics.
Episode | Topic |
1 | 100% fruit juices vs imposters; juice portion size |
2 | Real fruit vs fruit imposters |
3 | Real vegetables vs vegetable imposters |
4 | Recipe substitutions |
5 | Reinforcing portion size; memory joggers (eg, baseball, tennis ball) |
6 | FV for breakfast; breakfast on the go |
7 | FV in fast food restaurants |
8 | FV vs non-FV |
9 | Identifying number of FV in recipes |
10 | Review |
Mini-game descriptions.
Episode | Game | Description |
1 | 100% Fruit Juice | The player creates chains by linking together 100% juices, 100% juice blends, or Not Juices. |
2 | Find the Fruit | From the various foods floating in the bubbles, the player must pop the bubbles with the fruits. The player must also identify fruit imposters – ie, high fat items containing some or no fruit. |
3 | Find the Veggies | From the various foods floating in the bubbles, the player must pop the bubbles with the veggies. The player must also identify veggie imposters – ie, high fat items containing some or no veggies. |
4 | Lunch-a-Bunch | The player must add fruit and vegetable items to the passing lunch trays to create lunches with 2 servings of fruit and/or vegetables. |
5 | The Mole Pole 1 | A trivia game which tests the player’s knowledge of the information presented this episode; each correct answer allows them to progress through a tunnel. The goal is to exit the tunnel and get past the moles that guard it. |
6 | Breakfast Blunder | The player must add possible breakfast fruit and vegetable items to the food trays to create a breakfast with 2 servings of fruit and vegetables. |
7 | Fast Food Frenzy | The player must find possible fruit and vegetable choices on the menus of 3 different types of restaurants. |
8 | The Good Stuff | From the various foods floating in the bubbles, the player must pop the bubbles with fruit and vegetables. |
9 | The Mole Pole 2 | A trivia game which tests the player’s knowledge of the information presented this episode; each correct answer allows them to progress through a tunnel. The goal is to exit the tunnel and get past the moles that guard it. |
10 | The Mole Pole Review | This trivia game is structured similarly to the others; the difference is it tests knowledge presented from all previous episodes. |
Mini-game screen shot.
Decision making, led by the Wizard, occurred in episodes 1 - 10. In episode 1, players selected their top three personal values; for each value, they then selected “reason statements” that linked meeting their FV goals with each personal value they selected [
Goal setting, led by the Wizard, appeared in each episode. This component was tailored and interactive. In episode 1, players selected their favorite FV. In each episode, they were then presented with their favorite FV and decided which ones to use to meet their goal; they also selected the day(s) they would meet their goal. The FV goals became more difficult as the game progressed (
FV goals per episode.
Episode | Food | When | # of Days |
1 | F | Breakfast | 1 |
2 | F | Snack | 1 |
3 | V | Lunch | 1 |
4 | F or V | Dinner | 1 |
5 | V | Snack | 2 |
6 | F | Breakfast & Snack | 2 |
7 | V | Lunch & Dinner | 2 |
8 | FV schema | All day | 3 |
9 | FV schema | All day | 3 |
10 | FV schema | All day | daily |
Led by King Brocwell (the Good Guy) and King Sssynster (the Bad Guy), this component strengthened the player’s resistance to potential temptations [
This component appeared in episodes 1 – 4 to teach self-regulatory skills (ie, self monitoring, problem solving, asking/negotiation). This component was led by the game characters. Skills were taught through character modeling and dialogue.
Schemas are guides for complex behavior [
Schemas.
Character | Schema |
Knight Julie | 1B, 1L, 1D, 2S |
Knight Alex | 2B, 1L, 1D, 1S |
King | 1B, 0L, 2D, 2S |
Queen | 1B, 2L, 1D, 1S |
Wizard | 1B, 1L, 2D, 1S |
The virtual kitchen appeared in all ten episodes. It taught food preparation skills, planning, sequencing, and kitchen safety and promoted parent involvement. This component was interactive and included pre-steps involved in recipe preparation (ie, asking for permission, washing hands, etc) as well as a “virtual preparation” of the recipe (ie, a video clip that demonstrated how to prepare the recipe). With the exception of episode 1, players had a choice of recipes to prepare (
Recipes by episode.
Episode | Recipe Type | Recipe 1 | Recipe 2 | Recipe 3 |
1 | Juice | Razzle Dazzle Juicy Delight | n/a | n/a |
2 | Fruit | On-the-Run Trail Mix | Fantastic Fruit & Chocolate | n/a |
3 | Vegetable | Fiery’s Black Bean Burrito | M.I.C.H.A.E.L.’S Veggie Wrap | n/a |
4 | Fruit & Vegetable | Fivealot’s Famous Fruit Salad | Knight Brocwell’s Stuffed Potatoes | n/a |
5 | Vegetable | Knight Julie’s Veggie Snack | Fiery’s Bean Dip | n/a |
6 | Fruit | Royal Smoothie | Squire’s Strawberry Split | n/a |
7 | Fruit & Vegetable | Power Pudding Dip | Wizard’s Magic Pocket | n/a |
8 | Fruit & Vegetable | Celebration Sundae | Chef Supremo’s Cinnamon Carrots | n/a |
9 | Fruit & Vegetable | Queen Nutritia’s Dip | Platinum Sweet Potatoes | n/a |
10 | Fruit & Vegetable | Knight Alex’s Banana Pops | Golden Knight Burrito | Moledred’s Ice Pops |
The parent component consisted of newsletters and access to a parent website. Each is briefly described below.
Parent newsletters were designed to promote parent involvement (
Parent newsletter.
Episode | Focus | Recipe |
1 | 100% Fruit Juice | Peach Cobbler |
2 | Added Sugar in Fruits | Strawberry Shortcake |
3 | Vegetables | Black Bean Soup |
4 | Fruit and Vegetable Substitutions | Vegetable Lasagna |
5 | Serving Size Review | Tomato and Bean Dip |
6 | Vegetables for Breakfast | Breakfast Potatoes |
7 | Eating Out | Hearty Rice |
8 | Empty Calories | Round Table Pizza |
9 | Serving Size Comparisons | Vegetable Soup |
10 | Final Tips | Blueberry Dessert Cups |
Parent newsletter.
The parent website provided information to create a healthy home nutrition and activity environment. It included family-friendly recipes and addressed topics such as grocery shopping, eating on the go, and getting the family involved in physical activity. Information was routinely updated (ie, ten updates corresponding to the ten-episode video game) (
Parent website.
Episode | Focus | Recipes |
1 | Shopping Lists & A Well-Stocked Kitchen | Baked apples, cinnamon roasted sweet potatoes, spinach and strawberry salad |
2 | Getting Kids Involved in the Kitchen | Vegetable pasta, baked bananas, veggie grilled cheese |
3 | Family Meals | Spanish paella, gazpacho, and strawberry flurry |
4 | Family Activity Time | Bran muffins, vegetable omelet, fruit parfait |
5 | Buying & Storing Food | Chicken salad sandwich |
6 | Food & Kitchen Safety | Veggie couscous, turkey, light fish, pineapple orange frozen yogurt |
7 | Nutrition Facts Labels | Slow cooker chicken, three bean chili, tossed salad, potato bake |
8 | Portion Control | Layered salad, brown rice casserole, fruit salad, and chocolate berry cake |
9 | Eating on the Go | Fruit and nut mix, hummus with veggies, chicken salad sandwich |
10 | Substitutions | Broccoli mac-and-cheese, baked chicken nuggets, black bean brownies, pineapple angel food cake |
When children were eligible to play the next episode of the video game, an email with a link to the login page was automatically generated and sent to them. Simultaneously, parents received emails with links to the online newsletters and the parent website. An access database tracked participants (parents, children) through the program. Alerts notified the research team when parents and children were eligible to receive the next intervention component (ie, next game episode; next newsletter) and when they were eligible for data collection. The video game was programmed to notify the intervention staff when the child completed an episode, and emails were automatically generated when a parent opened the newsletter email or when the parent website was accessed. If the child did not log on to play the next episode of the game within approximately five days, they were contacted by the intervention staff. The intervention research team was available by email or phone to provide technical assistance. As part of process evaluation, all participant contacts were recorded in the access database.
To assess usual dietary intake, three unannounced 24-hour dietary recalls were obtained at each data collection period using the Nutrient Data System for Research (NDSR-2009), University of Minnesota [
Child psychosocial characteristics (FV preferences [
Following the framework of Baranowski and Jago [
Measures
Who | How | What | Baseline | During | Post 1a | Post 2b |
Child | Phone | FV intake (3, 24hr DR c) | x |
|
x | x |
|
Online | FV Preferences | x |
|
x | x |
|
|
FV Asking Behaviors | x |
|
x | x |
|
|
FV Self Efficacy | x |
|
x | x |
|
|
FV Outcome Expectations | x |
|
x | x |
|
|
Satisfaction With Change |
|
x | x | x |
|
|
Immersion |
|
|
x |
|
|
|
Social Desirability | x |
|
|
|
|
|
Game Likability |
|
|
x |
|
|
Gameplay | Logons |
|
x |
|
|
|
|
Responses/Choices (ie, goals set/attained, values/reasons, etc) |
|
x |
|
|
|
Interview | Game Reactions |
|
|
x | x |
Parent | Online | FV Intake (self) | x |
|
x | x |
|
|
Home FV Availability | x |
|
x | x |
|
|
Home FV Accessibility | x |
|
x | x |
|
|
Family Barriers | x |
|
x | x |
|
|
Self Efficacy | x |
|
x | x |
|
|
Child Asking Behaviors | x |
|
x | x |
|
|
Child Executive Function | x |
|
x | x |
|
|
Demographic Information | x |
|
|
|
|
|
Overall Reactions/Use |
|
|
x |
|
|
|
Email Open Rate |
|
x |
|
|
|
|
Website Visits |
|
x |
|
|
Staff | Logs | Process Evaluation | x | x | x | x |
a3 months after baseline assessment
b6 months after baseline assessment
cdietary recall
Repeated measures analyses of variance/covariance, controlling for key demographic factors, baseline FV consumption, and energy intake, accommodated a two-level within factor (post 1, post 2) and a four-level between-groups factor design. The group’s main effect allowed investigation of group differences, regardless of whether it was post 1 or post 2. The group-by-time interaction term allowed investigation of group differences over time, thus identifying if the treatment was maintained. Univariate outcomes (the number of goals achieved, number of newsletters read, etc) were analyzed using a univariate one between-group factor design. Secondary analyses included investigation of trends in goal attainment, FV consumption, and psychosocial factors across study weeks through the use of Chi-square analyses for ordinal repeated measures. A dose-response analysis was also planned.
This report provides a description of the protocol, procedures, and assessment tools for a video game designed to increase FV consumption among children. This research has several strengths. First, it is based on an earlier video game that successfully increased FV consumption in 4th grade children [
To our knowledge, this is the first study to test the effect of implementation intentions on FV goal attainment and consumption and to examine the relationship between satisfaction and maintenance of behavior change in pre-adolescents. The intervention includes a parent and child component designed within an integrated theoretical framework to maximize the likelihood of behavior change. The successful implementation of this intervention will generate valuable information regarding the effectiveness of this approach for young children.
Breakfast
Fruit(s), vegetable(s)
Lunch
Dinner
Not applicable
Snack
Squire’s Quest! II: Saving the Kingdom of Fivealot
This work is a publication of the United States Department of Agriculture/Agricultural Research Center (USDA/ARS), Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Texas. This project was supported by the National Institutes of Health, National Institute of Child Health & Human Development grant #HD050595 (to Dr. Thompson). This work is also a publication of the USDA/ARS, Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Texas, and funded in part with federal funds from the USDA/ARS under Cooperative Agreement No. 58-6250-0-008. The contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsement from the U.S. government. We would like to thank the youth and expert panel members who participated in this research. We would also like to extend our thanks to Archimage, Inc. of Houston, TX for its game design services.
DT was principal investigator of the study, conceived the project; guided the design of the intervention components; oversaw data acquisition, analysis, and interpretation; and drafted the paper. RB was with the Department of Pediatrics, Baylor College of Medicine when this study was conducted and assisted with design, managed the project, including intervention delivery and data acquisition; she also assisted with revising the manuscript. She is now with the Department of Family and Community Medicine, Baylor College of Medicine. ML was the Director of Marketing with Archimage, Inc (Houston, TX) when the Squire’s Quest! II game was designed and the executive producer for the Squire’s Quest! II game with responsibilities including project management, client communications, documentation, and copywriting; she assisted with design and participated in revising the manuscript. KC was a co-investigator and assisted with conception; dietary data collection, analyses, and interpretation; and revising the current manuscript. JB helped design and manage the original Squire’s Quest! study; she also participated in the design of the Squire’s Quest! II study and participated in revising the present paper. TB was the principal investigator for the original Squire’s Quest! study; he participated in the conception and design of the Squire’s Quest! II study and participated in analysis and interpretation; he also participated in revising the present paper. All authors read and approved the manuscript.
None declared.