Children’s vegetable intake is too low, and a key barrier to the inadequate intake is low acceptance. To facilitate successful development of new vegetable-based products for children, a sensory science approach to product development has been taken. A new theoretical model is proposed, the CAMPOV model: Children’s Acceptance Model for Product development of Vegetables. The model is informed by scientific literature and considers biological, psychological, and situational, and intrinsic and extrinsic product factors relevant to children’s acceptance of vegetables, with a focus on modifiable factors at the product level. Simultaneously, 14 new vegetable-based product concepts for children were developed and evaluated through focus groups with 5–8-year-olds (n = 36) as a proof-of-concept evaluation of the model. Children had high interest in six of the concepts. Factors identified from the literature that positively associated with the children’s interest in the concepts included bright colours, bite-sized pieces, good taste, fun eating experience, and familiarity. The CAMPOV model and proof-of-concept evaluation results can guide further sensory and consumer research to increase children’s acceptance of food products containing vegetables, which will in turn provide further insights into the validity of the model. The food industry can use the model as a framework for development of new products for children with high sensory appeal.

Children’s vegetable intake remains inadequate and school canteens may provide opportunities to address this public health concern. This study aimed to determine the effectiveness of an 8-week multi-strategy behavioral intervention that included vegetable provisioning and online menu architecture on vegetable sales in primary school canteens. A randomized controlled trial was undertaken in 16 Australian primary schools (n = 4302 students). The control arm kept their regular canteen menu. The primary outcome was vegetable sales measured by assessing vegetable content (in grams) from all menu items and using canteen sales (ordered online and over-the-counter) to calculate vegetable sales (in grams/week) at baseline (3 weeks) and during intervention implementation (8 weeks). Secondary outcomes were vegetable sales in subcategories, intervention acceptability among canteen managers and vegetable waste (four schools). Linear mixed model analysis showed that from baseline to follow-up, the intervention group had significantly higher weekly vegetable sales overall compared with the control group (2707 g/week, 95% CI 1276 to 4137 g/week; p < 0.001), with increased vegetable sales in the subcategories of burgers, hot foods and snacks, but not in sandwiches and pasta/rice dishes. The intervention did not lead to more vegetable waste, nor to a decrease in canteen revenue. The canteen managers found the intervention easy to implement and felt children responded favorably to three of the seven strategies. In conclusion, a multi-strategy behavioral canteen intervention increased vegetable sales amongst primary school students.

The teacher-led implementation of healthy eating programs in schools is cost-effective and potentially impactful. Teacher acceptability is important for uptake; however, process evaluations are scarce. This study evaluated the effect of two intensities of teacher training on the evaluation of a vegetable education program for Australian primary schools by teachers. The teachers (n = 65) who implemented the program as part of a cluster RCT (25 schools in two states, New South Wales and South Australia) received either low- (provision with materials and online training) or high (additional face-to-face (F2F) training)-intensity training prior to implementing a 5-week vegetable education program. They evaluated the acceptability of a digital training module and program by indicating the level of agreement with 15 and 18 statements, respectively, using 5-point Likert scales. The average item scores ranged from 3.0 to 4.2. All but one item, including student engagement, alignment to the curriculum and intent for reuse of the program, had a rounded average or median score of 4. The level of training intensity did not impact the teacher acceptability ratings. In conclusion, the teacher acceptability was good, and additional F2F training does not add value above the solely digital training of the teachers.

Children’s vegetable intakes are too low, and school canteens could provide an equitable environment to improve their intake. This study aimed to develop and apply a systematic method to quantify the proportion and variety of vegetable-containing items on primary school canteen menus and examine differences between schools of different socio-economic statuses, sizes and types. Online canteen menus from 112 primary schools in Sydney, Australia, were collected, and data were extracted on a total number of items and vegetable-containing items across different menu categories. Further, data on preparation type and variety were extracted. Differences in the proportion of vegetable items between socio-economic status, school size and type were tested. On average, 80.4 ± 34.0 items were listed, with 30% of items containing vegetables. Most sandwiches (60%) and hot foods (54%) contained no vegetables. The variety of raw vegetables (4.9 ± 1.8 types) was greater than for cooked vegetables (1.3 ± 1.2 types; p < 0.01). Limited differences were observed by socio-economic status and school type. Small schools offered fewer vegetable-containing items than large schools. While primary school canteen menus listed a large variety of items, only one-third contained vegetables. Data from this study can be used to track changes and to develop new opportunities to increase the vegetable supply in schools.

Interactions between Brassica vegetables and human saliva can affect in-mouth odor development, which in turn may be linked to individual perception and liking. S-Methyl-l-cysteine sulfoxide is a unique substrate present in Brassicas that produces odor-active sulfur volatiles due to the activity of enzymes present in plant tissue and due to bacteria, which may be present to varying extents in an individual's oral microbiome. Proton transfer reaction mass spectrometry was applied to measure individual differences in sulfur volatile production in real time when fresh human saliva was incubated ex vivo with raw cauliflower for a cohort of child-adult pairs. Large differences in the rate of sulfur volatile production were measured between individuals, but not between age groups. Significant positive relationships were found for volatile production between the adult-child pairs, suggesting a degree of commonality in saliva composition and oral microbiome activity. Furthermore, significant negative relationships were measured between the amount of in-mouth sulfur volatile production and liking for raw cauliflower in children.

The impact of different drying technologies (heat pump drying-HPD, freeze drying-FD, spray drying-SPD and refractance window drying-RWD) on tomato flavor was investigated using non-volatile parameters (pH, Brix, titratable acidity, glucose, fructose, and citric acid) and odor-active volatiles. Additionally, sensory differences were characterized between the pure tomato powders and as a flavor on potato chips. Relationship between volatiles and sensory attributes was elucidated by partial least squares regression and principal component analyses. Non-volatile components, except sugars showed significant difference (p < 0.05) between low temperature drying processes (FD, HPD) and commercial samples (SPD, RWD). HPD samples had the highest levels of fresh aroma volatiles compared to commercial samples: 1-penten-3-one (similar to 1.65 mg/kg dry wt.), hexanal (2.50), geranial (2.30) and beta-ionone (0.19) and were positively associated with higher fresh tomato flavor and sour taste. Thermally-induced volatiles; dimethyl sulphide (5.99), 1H-pyrrole-2-carboxaldehyde (0.22), 2-acetylfuran (0.23) and 3-methylbutanal (0.33) were present at higher concentrations in the SPD samples and were positively associated with tinned tomato flavor and bitter taste. The volatile compound methyl 2-methylbutanoate was associated (r = 0.61) with sweet taste. HPD was demonstrated to be a mild drying technology suitable for retaining fresh aroma volatiles and flavors of tomato and comparable to FD.

Schools provide a relevant and equitable environment to influence students towards increased vegetable consumption. This study aimed to evaluate the effectiveness of a Vegetable Education Resource To Increase Children’s Acceptance and Liking (VERTICAL) for Australian primary schools (curriculum aligned and based on a framework of food preference development and sensory experiential learning) on positively influencing factors predisposing children towards increased vegetable consumption. The secondary aim was to evaluate two levels of teacher training intensity on intervention effectiveness. A cluster-RCT amongst schools with three conditions was conducted: 1 = teaching VERTICAL preceded by online teacher training; 2 = as per 1 with additional face-to-face teacher training; 3 = Control. Pre-test, post-test and 3-month follow-up measures (knowledge, verbalization ability, vegetable acceptance, behavioural intentions, willing to taste, new vegetables consumed) were collected from students (n = 1639 from 25 schools in Sydney/Adelaide, Australia). Data were analyzed using mixed model analysis. No difference in intervention effectiveness was found between the two training methods. Compared to the Control, VERTICAL positively affected all outcome measures after intervention (p < 0.01) with knowledge sustained at 3-month follow-up (p < 0.001). In conclusion, VERTICAL was effective in achieving change amongst students in mediating factors known to be positively associated with vegetable consumption.