Validation of a food quantification picture book and portion sizes estimation applying perception and memory methods

Area of Science:

• Nutrition Science
• Food Consumption Studies
• Dietary Assessment Methods

Background:

• Accurate food portion estimation is crucial for dietary exposure assessment.
• Existing tools like the EPIC-SOFT food picture series require validation in diverse populations.
• Understanding perception and memory biases in portion size estimation is essential.

Purpose of the Study:

• To evaluate the Hungarian applicability of the EPIC-SOFT food picture series for dietary surveys.
• To investigate errors in food portion estimation stemming from visual perception and memory recall.
• To assess the accuracy of food portion estimation across different age groups.

Main Methods:

• Sixty-two participants across three age groups (10 to <74 years) were involved.
• Participants estimated portion sizes of five different foods presented in three sizes using the EPIC-SOFT series.
• Estimation accuracy was evaluated based on relative difference (≤25%) and relative standard deviation (<30%) after winsorisation.

Main Results:

• The EPIC-SOFT picture series was deemed acceptable for all five food items.
• Small portion sizes were generally overestimated, while large portions were underestimated.
• Memory recall led to significant overestimation, with errors up to 60.7% for certain foods.

Conclusions:

• The EPIC-SOFT food picture series is a viable tool for Hungarian food consumption surveys.
• Visual perception and memory recall significantly influence food portion estimation accuracy.
• Further refinement of estimation methods is needed to minimize biases, particularly for memory-based recall.