XML Schema 1.0 - qualified versus unqualified: what specifically changes in a valid instance? - validation

Question

Say I have a valid XML instance file ("1") that contains only elements in the targetNamespace of its schema (Schema "A"). The instance is valid per this schema. Instance 1 has the correct namespace declaration for its schema on its (1's) root node, and nowhere else.
Now I take the same instance 1 and try to validate it against Schema "B".
B is identical to A except I changed elementFormDefault="unqualified" to elementFormDefault="qualified".
I am seeing 1 failing to validate against B. Why? What do I need to change in 1 (producing instance "2") to make it valid again against B?
Is the "qualified" flavor of an XML schema just for instances with an explicit namespace prefix on every single element?
Example schemas: All based on a minimal XML structure for communicating some data structures like file/record/info1. Unfortunately these seem to behave differently again from my real-world examples (which are too big to post here).
A: UNqualified schema
<?xml version="1.0" encoding="UTF-8"?>
<!-- The UNqualified version -->
<xs:schema elementFormDefault="unqualified" targetNamespace="http://example.com/xsd-prefixes-test"
xmlns:test="http://example.com/xsd-prefixes-test" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<xs:element name="file">
<xs:complexType>
<xs:sequence>
<xs:element maxOccurs="unbounded" ref="test:record" />
</xs:sequence>
</xs:complexType>
</xs:element>
<xs:element name="record">
<xs:complexType>
<xs:sequence>
<xs:element ref="test:info1" />
<xs:element ref="test:info2" />
</xs:sequence>
<xs:attribute name="id" type="xs:integer" use="required" />
<xs:attribute name="status" type="xs:NCName" use="required" />
</xs:complexType>
</xs:element>
<xs:element name="info1">
<xs:complexType />
</xs:element>
<xs:element name="info2">
<xs:complexType />
</xs:element>
</xs:schema>
B: QUALIFIED schema
<?xml version="1.0" encoding="UTF-8"?>
<!-- The QUALIFIED version -->
<xs:schema elementFormDefault="qualified" targetNamespace="http://example.com/xsd-prefixes-test"
xmlns:test="http://example.com/xsd-prefixes-test" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<xs:element name="file">
<xs:complexType>
<xs:sequence>
<xs:element maxOccurs="unbounded" ref="test:record" />
</xs:sequence>
</xs:complexType>
</xs:element>
<xs:element name="record">
<xs:complexType>
<xs:sequence>
<xs:element ref="test:info1" />
<xs:element ref="test:info2" />
</xs:sequence>
<xs:attribute name="id" type="xs:integer" use="required" />
<xs:attribute name="status" type="xs:NCName" use="required" />
</xs:complexType>
</xs:element>
<xs:element name="info1">
<xs:complexType />
</xs:element>
<xs:element name="info2">
<xs:complexType />
</xs:element>
</xs:schema>
Example schema instances:
Instance 1 - prefixes for the targetNamespace on all elements; validates against both schemas
<?xml version="1.0" encoding="UTF-8"?>
<!-- ALL prefixes -->
<test:file xmlns:test="http://example.com/xsd-prefixes-test"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://example.com/xsd-prefixes-test xsd-prefixes-test_qualified.xsd">
<!-- validates against xsd-prefixes-test_UNqualified.xsd AND xsd-prefixes-test_qualified.xsd -->
<test:record id="1" status="ok">
<test:info1 />
<test:info2 />
<!-- etc... -->
</test:record>
<test:record id="1" status="ok">
<test:info1 />
<test:info2 />
<!-- etc... -->
</test:record>
<test:record id="1" status="duplicate_deprecated">
<test:info1 />
<test:info2 />
<!-- etc... -->
</test:record>
</test:file>
Instance 2: - no namespace prefixes anywhere but still validates against A and B!
<?xml version="1.0" encoding="UTF-8"?>
<!-- NO prefixes anywhere -->
<file xmlns="http://example.com/xsd-prefixes-test" xmlns:test="http://example.com/xsd-prefixes-test"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://example.com/xsd-prefixes-test xsd-prefixes-test_qualified.xsd">
<!-- validates against BOTH xsd-prefixes-test_UNqualified.xsd AND xsd-prefixes-test_qualified.xsd -->
<record id="1" status="ok">
<info1 />
<info2 />
<!-- etc... -->
</record>
<record id="1" status="ok">
<info1 />
<info2 />
<!-- etc... -->
</record>
<record id="1" status="duplicate_deprecated">
<info1 />
<info2 />
<!-- etc... -->
</record>
</file>
Instance 3 - does not validate against either A or B - why not?
<?xml version="1.0" encoding="UTF-8"?>
<!-- NO prefixes anywhere -->
<test:file xmlns:test="http://example.com/xsd-prefixes-test"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://example.com/xsd-prefixes-test xsd-prefixes-test_qualified.xsd">
<!-- validates against NEITHER xsd-prefixes-test_UNqualified.xsd NOR xsd-prefixes-test_qualified.xsd -->
<record id="1" status="ok">
<info1 />
<info2 />
<!-- etc... -->
</record>
<record id="1" status="ok">
<info1 />
<info2 />
<!-- etc... -->
</record>
<record id="1" status="duplicate_deprecated">
<info1 />
<info2 />
<!-- etc... -->
</record>
</test:file>
Error messages (same with both schemas A and B):
Instance 3 - cvc-complex-type.2.4.a: Invalid content was found starting with element 'record'. One of '{"http://example.com/xsd-prefixes-test":record}' is expected.
I am seeing this error or very similar with instance files along the pattern of both 1 and 2, when used with the schema of the "opposite" flavor (an instance with prefixes against a schema with unqualified).
Conversely I am seeing instances of style 3 that do validate against the qualified type schema.
Is there a dependency on what language/parser/validation engine is used?