For many families, mealtimes can become one of the most difficult parts of the day. Some autistic children and adults accept only a very limited range of foods, reject certain textures, cannot tolerate particular smells, or need food to be presented in a very specific way.

Sometimes these behaviors are dismissed as being “picky,” “spoiled,” or simply refusing to try something new.

However, a new scientific review reminds us of something very important: for many autistic people, eating is also a sensory experience.

A study published on October 3, 2026, in the Journal of Autism and Developmental Disorders brought together findings from 46 studies on feeding and sensory processing in autistic children. Of these, 17 were included in a meta-analysis.

The researchers found a consistent relationship between differences in sensory processing and feeding-related difficulties.

This does not mean that every eating difficulty experienced by an autistic person is caused by sensory factors. But it can help us understand why simply asking someone to “just try a little” may not be as easy as it sounds.

Eating involves many senses at the same time

When we think about food, we often think first about taste.

But eating involves many other senses as well.

There is the texture of food in the mouth, its temperature, smell, appearance, the sound it makes when chewing, and even the way it feels on the lips, tongue, or hands.

For someone with sensory processing differences, some of these experiences may feel much more intense.

A texture that another person barely notices may feel unpleasant or very difficult to tolerate.

A smell may feel overwhelming.

A crunchy food may create a completely different sensory experience from a soft food.

Even two foods that taste almost the same can feel entirely different because of their texture.

What did the new review find?

The researchers found that autistic children experienced more mealtime difficulties than non-autistic children in the comparison studies.

The most noticeable sensory differences appeared in the tactile domain, but differences were also found in areas related to:

• taste and smell;

• vision;

• sound;

• and other sensory experiences connected with eating.

When the available results were combined, the researchers also found a moderate association between sensory processing differences and feeding difficulties.

In other words, greater sensory processing differences tended to be associated with greater feeding-related difficulties.

It is important to remember that an association does not necessarily mean that one directly causes the other.

What looks like “refusal” may have an explanation

This kind of research can help us rethink how we interpret certain behaviors.

Consider a child who only accepts crunchy foods.

Perhaps they are not trying to be difficult.

Crunchy textures may feel predictable, while soft or mixed foods may create sensations that are very uncomfortable for them.

Another person may tolerate one specific brand of yogurt but refuse another that seems almost identical because the consistency is slightly different.

A food may also be accepted at home but rejected somewhere else because its temperature, presentation, smell, or even the noise level in the environment has changed.

For those of us who do not experience these sensory differences, it can be difficult to understand how something that seems small can completely change the experience of eating.

That is why it is so important to observe before judging.

For autistic people, these experiences are real

It is also important to listen directly to autistic people.

Many autistic adults have described for years how strongly they experience food textures, smells, temperatures, and presentation.

We should not assume that someone is exaggerating simply because we do not experience the same sensations.

A sensory response can be very real even when other people cannot see it.

Understanding this can completely change the way we support someone.

Instead of asking:

“Why won’t they eat it?”

perhaps we can begin by asking:

“What might they be experiencing when they try to eat it?”

The difference between those two questions is significant.

Forcing someone to eat may not solve the problem

When a person persistently refuses certain foods, families are often given advice such as:

“If they are hungry enough, they will eat.”

“Don’t give them anything else.”

“They need to get used to it.”

But if a significant sensory difficulty is involved, forcing someone to tolerate an experience that feels extremely unpleasant may increase anxiety around food.

The goal should not be to turn every meal into a battle.

This also does not mean that we should give up on expanding someone’s range of foods when there are nutritional concerns.

It means that we should first try to understand what is making eating difficult.

For some people, it may help to introduce changes gradually.

For example, the first step may simply be tolerating a new food on the plate, then touching it, smelling it, and only later trying a small taste.

But every person is different.

What works for one person may not work for another.

Therapists can also look beyond the behavior

This study also carries an important message for professionals.

When feeding difficulties are present, it is not enough to look only at whether a person “eats” or “doesn’t eat.”

It may be important to explore:

• which textures they accept or reject;

• which temperatures they tolerate;

• whether certain smells trigger discomfort;

• whether they prefer foods mixed together or kept separate;

• whether noise in the environment makes eating harder;

• whether there are difficulties with chewing or swallowing;

• whether pain, constipation, reflux, or other medical issues are present;

• and how the autistic person themselves feels about food and mealtimes.

In some cases, support from professionals such as occupational therapists, dietitians, feeding specialists, speech-language pathologists, or physicians may be appropriate, depending on the specific concern.

Whenever possible, the autistic person should be included in those decisions.

We should avoid another mistaken conclusion

Although this review found an important association, we should not go to the opposite extreme and assume that every feeding difficulty in autism is sensory.

It is not.

A person may experience difficulties with eating for many different reasons, including gastrointestinal problems, allergies, oral-motor difficulties, anxiety, previous negative experiences with food, medication side effects, or other medical conditions.

That is why professional assessment is important when food intake becomes extremely limited, when there is weight loss, nutritional deficiency, pain, difficulty swallowing, or a sudden change in eating patterns.

A change in perspective can make a big difference

For me, one of the most important lessons from this kind of research is not only found in the numbers.

It is found in the way it encourages us to look at the person.

When we understand that a behavior may be connected to a different sensory experience, we stop immediately thinking:

“They are being difficult.”

And we begin to ask:

“What do they need?”

That does not mean removing every challenge or giving up on teaching new skills.

It means providing support in a way that is more informed and respectful.

Autistic people do not need us to ignore their difficulties.

They need us to try to understand them.

And when it comes to eating, understanding how someone experiences tastes, textures, smells, and the environment may be the first step toward finding strategies that can truly help.

At Positive Autism, we believe that supporting a person begins with trying to understand their experience, not forcing them to behave the way we expect them to.

Scientific source

Szalkay, K. and colleagues (2026). Early-Life Feeding Behavior and Sensory Processing in Autism Spectrum Disorder: A Systematic Review and Meta-Analysis. Journal of Autism and Developmental Disorders. Published October 3, 2026.

The review included 46 studies, with 17 included in the meta-analysis.

https://link.springer.com/article/10.1007/s10803-026-07543-6?utm