Learn English – the origin of “rings a bell”

According to Etymonline:

To ring a bell "awaken a memory" (1934) is perhaps a reference to Pavlovian experiments.

Here's an excerpt from that last linked page:

Pavlov became interested in studying reflexes when he saw that the dogs drooled without the proper stimulus. Although no food was in sight, their saliva still dribbled. It turned out that the dogs were reacting to lab coats. Every time the dogs were served food, the person who served the food was wearing a lab coat. Therefore, the dogs reacted as if food was on its way whenever they saw a lab coat.

In a series of experiments, Pavlov then tried to figure out how these phenomena were linked. For example, he struck a bell when the dogs were fed. If the bell was sounded in close association with their meal, the dogs learnt to associate the sound of the bell with food. After a while, at the mere sound of the bell, they responded by drooling.

Another possible origin is the one this page advocates:

There are many bells that ring to remind or instruct us to do things: doorbell (open the door), telephone bell (pick up the phone), school bell (come to class), toaster bell (take out the toast), and the clothes dryer bell (take out the clothes). So, if something such as a face, a name, a number, or a date "rings a bell," it causes you to remember something.

To counter the tiresomely inaccurate Google Ngrams, I've decided to do some research using the COHA corpus:

The first idiomatic usage listed dates from 1941:

Next time they hear the name Sammy Glick it rings a bell in their brains.

I googled the phrase, and the only result (Google books) was the link to Budd Schulberg's 1941 short story.

To my way of thinking, the in their brains is significant in determining the age of the phrase. It sounds a lot like the phrase was new at the time, since the writer felt the need to explain it a little.

By the next usage listed, in 1943, rings a bell was used as it is today, with no quantifier phrase. After these instances, the idiomatic usage occurs often.

Another point in the Pavlov theory's favor is the fact that the number of occurrences jumps after the 1930s. I tested the corpus by searching for the word "the":

While 1810 and 1820 had much smaller numbers (69742 and 433985), the number for each decade resides pretty consistently between a million and one and a half million instances. This shows the extent of the corpus.