Meta-learning is a powerful framework that enables machine learning to acquire the ability to learn new tasks quickly and efficiently. In this paper, the application of meta-learning to the field of psychological needs prediction has been explored. Unlike most of the current machine learning models that are ‘task specific’ and require huge amounts of training data, the work described in this paper is able to do few-shot learning for a new learning task. The Model-Agnostic Meta-learning (MAML), a Meta-learning algorithm, has been used in this work to train a model capable of rapidly generalizing to different tasks with a very limited amount of labeled data. This work attempts to overcome the limitations of traditional machine learning approaches in the field of psychological needs prediction and is shown to perform well in its few-shot learning approach.